Module 1.1: Administrivia & Intro (Part 1) | Data Mining& Warehousing (2025/26 CSC 4792)

Added: 2026-05-25

[00:01:58]Brains.

[00:02:13]>> All right. Morning, everybody.

[00:02:16]Checking to see if uh if you can hear me.

[00:02:23]Hello. Can you hear me?

[00:02:37]Checking. Hello. Can you hear me?

[00:02:43]I don't know if it's my connection here.

[00:02:45]I see some reactions, but uh I'm not quite sure if these are trusting.

[00:03:00]Yes, sir. We can hear you. Thank you.

[00:03:08]Yeah. So, I just realized something. I realized that um >> [clears throat] >> it appears there's uh there's a bit of an issue here. Um and I'm not really sure if this is going to work out here. There's a bit of an issue in terms of uh Um so, I'm using Google Meet, right?

[00:03:29]And uh it turns out that it turns out that I think there's a limit.

[00:03:36]I was actually interacting with um somebody from from our CICT.

[00:03:40]And I'm told there's a No, I'm told actually there's a There should be a limit in terms of how many people um can actually be enrolled into into a session, a single session.

[00:03:54]Uh and so what I have done is uh what I've done is I've uh I'm attempting to stream this also.

[00:04:01]Uh seeing as this is an admin stream here.

[00:04:05]I don't know why this is the case here.

[00:04:07]So, I'm not really sure if uh the stream is is working just fine. I hope it is.

[00:04:14]If it is not, then I don't know. We're going to have to But, I'm also recording the uh what do you call this? The session.

[00:04:22]So, I'll ask um I have no idea if we if we have um course representatives yet, but I'll ask uh you guys to probably uh share the stream link for those that uh are going to be unable to to join.

[00:04:39]>> [clears throat] >> I've just shared the the link, but what I'll do is I'll also uh I'll also send uh send an email.

[00:04:48]Um apologies for that. Um Yeah, but but hopefully this this issue will be addressed before we go full swing here.

[00:04:59]Um All right, on with it. Um I'm I'm I'm hoping you can also see my uh my my screen. I think you can because I can see it on my mobile device.

[00:05:15]Um Yeah, these are these are really really interesting times we're living in. You know, I was I was I was looking up uh there's a number of conversations that have been going on regarding you know, online uh online classes and and things of that sort.

[00:05:32]Um and it turns out that uh these conversations and debates actually have have been taking [music] place also amongst uh amongst members of There we go.

[00:05:43]Um Anyways, so so my my advice to you, right, is um um you want to you want to keep an open mind.

[00:05:55]I I know your your student union has has issued a number of statements. Um I guess for the most part trying to um Well, I say trying to uh discredit here, but but they they're seemingly presenting arguments for why um for why the university shouldn't really reopen, why the university should not have reopened, if you will.

[00:06:21]But but anyways, um we are just going to go with the flow, you know, try as much as possible to see if we can figure out what will work best for for us, I think.

[00:06:34]Uh so, initially, what we're going to do is we're going to do a bit of experimentation. And I think you'd have you'd have noticed uh this from the very very strange timetable, right? We have a 12-hour session and 1-hour long session or something like that. Um it's not cast in stone. If it turns out that uh that what we are doing right now is is is not going to be workable, we we will change things slightly. At at at least uh so we can buy ourselves some time for when we we are all going to return to campus, I guess, uh physically. Anyways, so so we're going to start here, um everybody. Uh just going to see if I can You know, OBS is uh chewing up a lot of my It's it's acting up here, and I I don't know what this is all about, yeah.

[00:07:29]It's messing up all my my things here.

[00:07:34]Sorry about that. The streaming is what's causing all of this, by the way.

[00:07:39]It's not I think it's the streaming, guess.

[00:07:43]And I hope I am streaming. I don't know if I'm streaming.

[00:07:49]Let's see.

[00:07:51]There's only one sure way to find out if we are streaming, is it? We can we can probably check right now.

[00:07:58]Um yeah, so so we're going to start here. Um if if if um something is going to appear off uh if you're uncomfortable about the way things that are done or something, we can always have a discussion and try to get up here and I I don't know why. Okay, so I think we are streaming. We can always have the a discussion um and try and alter things slightly, you know? Uh yeah, so I see a message here. Miss Miss Nyirenda, I hope I'm hoping that you're going to I don't know if probably you're going to be chosen as uh the course rep for this course as well.

[00:08:38]Um so those that are unable to join, it's it's a limitation on Google Meet.

[00:08:42]I'm using Google Meet. So, what I I've done is I've sent a a direct link via via email. I've also posted it right now.

[00:08:51]Just share that link to them and they'll be able to join the stream.

[00:08:56]Thank you very much.

[00:08:58]Um Thank you very much.

[00:09:02]Let me just see if I can admit all of them or something.

[00:09:09]It's very strange that I've I've I've activated auto admit, but uh this is seemingly not working, I guess. I don't know.

[00:09:16]Uh No, why this thing also is telling me you lost network connection. Try. All right. So, anyways, um on with it. Uh I guess we can we can kind of start here.

[00:09:42]You can start.

[00:09:53]I have no idea what's happening with my my connection here. That's okay.

[00:09:58]I mean my my computer here says fine.

[00:10:18]All right. There we go.

[00:10:19]Okay. So so on with it. So welcome everybody to um to CSC 4792.

[00:10:27]Uh so CSC 4792 is officially called data mining and warehousing.

[00:10:32]Pretty interesting course, I think.

[00:10:35]Um Same here in part because I've been I've been teaching uh this course at postgraduate level for a very very long time.

[00:10:44]Um and and I quite quite enjoy teaching the course.

[00:10:48]Um I I know I I taught some of you in 4505 uh in in computer graphics and visual computing, but I just realized that the sheer size of students has actually increased. So I'm going to have to reintroduce myself, I suppose.

[00:11:04]So my name is Alaiton Terry.

[00:11:06]Okay. Um I assume you are lecturing in the Department of Computing and Informatics.

[00:11:11]Um and yeah, so pretty excited to to be [clears throat] doing this with you. So today's interaction is going to be uh it's going to be a somewhat short, right? It's um meant to introduce us to the course, obviously, and more importantly to to set the stage for how exactly the course is going to be administered ideally. So we're going to spend the the a significant part of of this morning discussing admin trivia, right?

[00:11:39]Um and and in essence, uh these are the things that we're going to discuss. We want to make sure that we're on the same page.

[00:11:47]Um almost always happens somewhere towards the end of the course, um some people pop out of nowhere.

[00:11:53]"Oh, I didn't submit this assignment."

[00:11:55]Or uh "I did not know that there were marks being directed to this particular assessment component or something like that." Right? So, we're we're we're trying to make sure that we are all on the same page, yeah?

[00:12:06]Um if you're wanting to um find out a bit more about me, uh there's a lot that goes on in the department. Um uh there's a there's a research group that I I founded. I'm also principal investigator there. It's called the DataLab Research Group.

[00:12:23]Um so, if you go to datalab.unza.zm/people, you will find uh details about myself and um students, past students, and current students, and also colleagues that I collaborate with. You know, okay?

[00:12:37]Um and really what we do in our lab um is we we we conduct research, obviously.

[00:12:44]So, research, consultancy, and community engagement. And and everything we do, all these activities are actually aligned to three broad computing areas or subfields.

[00:12:56]So, data mining, which should be of interest because it's it's a it's a key part of what we're doing in CSC 4792.

[00:13:05]Um we also do quite a bit of uh um uh research in subfield of computing called digital libraries. It's actually uh a a fancy way of referring to data management, right?

[00:13:18]Ideally, yeah?

[00:13:19]Um >> [clears throat] >> but but the the former subfield is referred to as digital libraries in computing.

[00:13:25]Um and then because of my my my background, uh, what I did, what I focused on during my doctoral studies, we also conduct research in subfield of computing core technology in East London. Really, this is, uh, again, a fancy term, um, um, aligned with, uh, educational technology, so education technology.

[00:13:47]Okay. Uh, so anything related to each and every one of these subfields. Now, if you think about it, you begin to realize that, uh, uh, this is still at a very high level, right? Um, if you go to our research group website, and especially, uh, the the publications page, um, you you begin to realize that we do quite a bit there.

[00:14:07]Um, a few highlights, uh, of the projects that we've we've been working on, that we're working on, um, as a lab.

[00:14:15]Um, there's a road safety um, there's a road safety, uh, in Zambia project that was initiated back in, is it 2023 or 2024? Some some something like that.

[00:14:27]Um, where we are essentially interested in trying to figure out how computing can be leveraged, right? To to to to to improve road road safety in Zambia.

[00:14:39]And in the recent past, we've we've actually focused on um, trying to determine how best we can take advantage of advances in machine learning and artificial intelligence to actually achieve this goal objective.

[00:14:55]You know, so I currently have students that are, you know, working on on interesting problems where they are mining specific types of data, the images, so images of road controls, they're called, right?

[00:15:08]So, road signs, um, and really the state of roads, and and trying to do fancy things like trying to um, to to to to determine, um, anomalies as they exist on the roads, I suppose.

[00:15:22]Um, and then back in 2024, I think it was, I worked with a group of students who worked towards uh building um uh a a a pothole detection kind of end-to-end solution, if you will.

[00:15:37]Okay.

[00:15:40]Anyways, um and then we've also done, you know, work uh as it relates to um accessibility uh when it comes to legal and legislative documents.

[00:15:52]You know, um the pre-LLM era led us down the path where we're trying to experiment with what's typically called um uh is it abstractive and extractive summarization, right? These are the key um uh techniques used when it comes to summarization of of large documents, right? I'm saying pre-LLM era here because uh it turns out that with with with advent of these so-called foundation models, this this this um this problem is virtually solved, right? You don't really need to bang your head against the wall to to try and figure out how best you can do this. So, what we're trying to do really is try and um uh we're solving a simple problem, right? If you look at your typical um legislative document, um it's it's difficult for a layperson to actually understand, right? To read and understand the document. Uh in part because these these documents are long documents, they're typically lengthy, and they're also too technical, right?

[00:16:51]So, you you you wouldn't really be able to understand everything that is in that document. And so, what we did was we we set out to try and see if we could build software that would be able to, you know, kind of like summarize these documents, right?

[00:17:05]Um make it a lot easier for the average person in Zambia to be able to understand these documents.

[00:17:11]Um we've also been doing quite a bit uh as it relates to educational data mining.

[00:17:17]Um and I like using um uh a certain example because uh especially for this course because uh uh I'm actually we're actually going to use it as a case uh a case study, if you will.

[00:17:31]You know, so our interest when it comes to educational data mining is to try and figure out exactly how we can take advantage of data that is generated in educational settings.

[00:17:41]Right? So, as we speak right now, right? As we speak right now, there's uh there's data being generated, right? We We have a whole slew of people that uh um you know, currently logged on.

[00:17:54]Uh I'm not really sure if there's any other interesting information that we can potentially um I guess extract from Google Meet here, but probably just uh um uh attendance logs, I guess, right? I mean, if we wanted to make this fun, so we could probably uh take advantage of uh these tools that exist in Meet, right?

[00:18:16]Like, do polls, for instance. Is it We can create polls.

[00:18:21]And uh try and see if we can we can start collecting more nuanced information or data, ideally. And as you say, an example of like uh data as it relates to educational data mining, right? Um but also, it turns out that there's uh there's uh there's there's there's other interesting data that you can collect from platforms like the Moodle, right? So, anyways, so so we we we also do The long and short of it is that we we we we've been doing, you know, quite a bit as it relates to educational data mining, right? Uh in essence, trying to to to figure out exactly how we can make uh teaching and learning more effective by understanding what people do in these sort of environments, right? Um by way of mining data in these environments.

[00:19:09]Uh and there's a a of places where you can mine data, right? It's not just moodle or here and stuff. There's a lot of platforms where you can mine this information.

[00:19:17]Um uh you know, SIS and all those things.

[00:19:21]Um and then in addition to that um I I have a personal interest as it relates to Wikipedia. Um there's there's another case example that I would want us to discuss in this course.

[00:19:36]Um there's a project a long-standing project we've been working towards trying to see um what it is we can do to um to increase you know, content contribution in the global south by people that reside in these sort of places, areas, right? So, if you look at Zambia right now, um and I like using an example of uh the University of Zambia Wikipedia page.

[00:20:00]Uh if you go there, you notice that it's uh it's a stale page, isn't it? In fact, it qualifies to be called a stub, right?

[00:20:06]Because it's incomplete. It has incomplete data.

[00:20:09]You know, um worst you if you do an analysis and try and figure out um uh which individuals, right? Or Wikipedians have actually been updating content, you begin to notice that uh it's largely people from outside Zambia, right? So, we've been doing quite quite a bit of things um as it relates to that. Um I've worked with a number of students that have have built tools, nothing to do with data mining, but they've built tools um that are aimed at uh uh increasing you know, content contributions from people uh residing in the global south, ideally. Um beginning is it last year, I think it was, I started working with uh postgraduate student, Fraser.

[00:20:50]Um Fraser is uh looking at uh uh leveraging the the the the the potential of LLMs, right? As it relates to content generation.

[00:21:03]Um and so, he's experimenting with what's called RAG. I'm sure if people have have heard about this, um so so the premise behind what he's doing is is quite simple, right? Um with with chatbots or LLM LLMs or chatbots like uh Gemini or ChatGPT, if let's say you wanted to generate content on and about Wonsa, right? You just probably go to um Gemini, right?

[00:21:29]Or perhaps Deep Seek, is it?

[00:21:30]chat.deepseek.com or something like that. And then you say, "Generate content for the University of Zambia Wikipedia page." The problem with that is that what you're doing is you're largely making use of what's called synthetic data. And it turns out, right?

[00:21:42]For people that have done studies in this space, it turns out that um the the internet is now riddled The World Wide Web is riddled with synthetic data, slope, they call it, right?

[00:21:55]Um this content that is largely generated by machines. Um and and unfortunately, um as you probably have all noticed, uh it's difficult to trust content that is generated by machines, right? Because these machines hallucinate, is it?

[00:22:13]They're designed They're token generators, right? If you look at the science behind LLMs, they're just nothing more than uh token generators, right? They start behind the scenes.

[00:22:22]Um and I hope I still have my connection.

[00:22:24]Here we go. Couldn't see myself on my on my mobile device.

[00:22:29]Um uh no, I'll pause here. Can you Can you hear me? Can you still hear me?

[00:22:36]Uh Hello, can you hear me and can you see my screen? Okay, thank you very much.

[00:22:41]And what's happening with my mobile device? I can't uh see myself here.

[00:22:45]Um all right, I can't see the It's you know, forgot how to go back. Okay. I couldn't see the um Uh that's fine. I I couldn't see the um um I couldn't see the screen, anyways.

[00:22:59]Uh anyway, so so so what Phrase is doing is is trying to leverage the power of foundation models, right? Um coupled with what's called retrieval limited generation, right? So this notion that uh you can actually force these LLMs to be accurate at what they do by giving them uh or uh giving them or feeding them a knowledge base, is it?

[00:23:21]Right?

[00:23:22]Uh if that makes sense.

[00:23:24]Okay. Um and then for the longest of time we've we've actually been um mining prior research output. It's scholarly research data if you will, right? Um so we we've been trying to figure out ways of improving um the or increasing the visibility of research output. Um uh the visi- visibility of research output uh in the global south. Uh particularly Zambia in this particular case, right?

[00:23:53]So quite quite a bit of stuff we've been um we've been doing there as well. Um and again uh this the data set we may we may uh end up using, I believe.

[00:24:06]Um Try and see here if there's uh maybe a bit bit of time to Yeah, hold on. Let me try that.

[00:24:16]To to to to use some of these data sets, right? And my my approach is always simple to this course. I prefer to use uh data sets that I've had a I've had a hand in building in in part because most of these data sets are actually uh aligned with problems that afflict like Zambia. And my my thinking is it's a lot easier for you to understand what we're doing if we're using uh locally created data sets, right? Can easily relate to these things.

[00:24:42]Especially the students at preschool failing kind of thing there.

[00:24:45]But anyways, uh so there's there's there's that. So we've done quite a bit here. And and in fact when it comes to the Zambia National uh ETD portal project here, we've built a number of models here, right?

[00:24:57]NLP models and also classifiers as well.

[00:25:00]Uh binary classifiers, you know, multi-class classifiers and multi-label classifiers also. Uh so, there's that.

[00:25:06]Um and then we've um we've been since 2022, we've been uh collaborating with uh colleagues from the University Teaching Hospitals.

[00:25:16]And uh as of is it last year, I think it was uh colleagues um uh colleagues uh from >> [clears throat] >> from um I want to say from uh the Cancer Diseases Hospital. Yeah.

[00:25:30]You know, uh so so our goal here is to uh to figure out we've been trying to figure out exactly what it is we can do to to uh ensure that radiological workflows are more efficient and effective. So, quite a bit we've been doing here.

[00:25:46]And as it relates to what we're doing, I guess mining data, the data we've been mining are images, right?

[00:25:53]Um and and in fact most of the things we've been doing have been restricted to a specific um category of images. These are called modalities. Anyway, um X-rays.

[00:26:04]Right? So, there's a whole slew of of medical images out there, right? MRIs, uh CTs, you know, ultrasound.

[00:26:12]Um and all those things. Um And also so the students I've been working with um and I'm happy to say one of the uh master students I've been working with since 2023 is going to be graduating in the next coming weeks.

[00:26:27]Uh did some amazing work. Um He built um he built a model um I guess it is.

[00:26:36]Built um an um uh a a He was working towards building Well, this will sound simple really, but it's two models, right? So, he he was working on what's typically called a a classification with localization problem, right? So, if you look at your typical and I explained this as as I introduced the course, I think there's a part where we um I get to showcase demonstrate some of the things that we have done in in great detail, right? I'm just trying to give you context in regards to what it is we're doing the lab here. Um but but but what he did was he went towards building two models, right? This is a design models aimed at uh uh helping um identify pneumonia cases ideally, community-acquired pneumonia as it were.

[00:27:23]Um so so what he did was he he built a um localization model, right? Uh and what he does is he does something simple. Given an input DICOM file, um it's a chest X-ray.

[00:27:35]What the model does is it draws bounding boxes, yeah?

[00:27:40]Highlighting problematic areas in that uh chest X-ray.

[00:27:44]Okay? Um and then he also built um a binary classifier which usually is probably one of the simplest possible things you can do, right? And given a DICOM file as input, what the classifier will tell you is whether the DICOM file or the X-ray image is associated with pneumonia or not.

[00:28:03]Um and it also spits out some um some some figure, the accuracy, right? Or the prediction itself, and the precision and recall and stuff like that. Anyway, so so so there's that.

[00:28:15]Um Now, it turns out that what we're going to be doing between now and I don't know if it's August here.

[00:28:22]Um I'm actually very happy that we started early because we won't really have to uh try and squeeze things into 12 weeks like we did last time. Okay? Um but anyway, so so the learning outcomes for CSE 4792, and incidentally they're somewhat aligned some of the things I've spoken about is we we're going to be working towards uh sort of situation where we get you to a stage where you'll be able to explain your data mining and your data warehousing concepts.

[00:28:50]Okay? The applications and associated challenges.

[00:28:55]If that makes sense.

[00:28:59]I don't know if I'm still streaming here. I can't see that.

[00:29:01]Um and then we want to make sure that you're able to process data, right? So, we will discuss this whole notion of data pre-processing. So, how do we go about cleaning data?

[00:29:14]Um cuz it turns out that when when you are when you harvest data, right? When you collect the data, usually it will be riddled with a lot of errors and anomalies, right? So, you want to clean it up or you process it ideally.

[00:29:28]Okay?

[00:29:29]Uh I'm going to pause here. Can you still hear me?

[00:29:34]Am I still audible? Good, thanks.

[00:29:37]Um >> [clears throat] >> Right. Um and then we we will also have a discussion of how exactly you go about preparing data, right? So, we'll look at various transformation techniques.

[00:29:49]Um and then we'll look at whole notion of dimensionality reduction, right? So, when you're working with a data set that has a lot of attributes, you know, how do you go about identifying the more important aspects of of the data itself, right? Um and then we are supposed to introduce you to how you go about implementing I don't want to call these data mining, you know, algorithms really because we're not going to be teaching you how to implement K-nearest neighbor, decision trees, random forest, and support vector classifiers. No.

[00:30:29]Our goal is to to show you exactly how to use these estimators, right? So, I want to say what we're going to be doing here is implementing data mining models, not algorithms, right?

[00:30:39]Because the models themselves are encapsulation of data.

[00:30:43]And um um uh and this meters with the algorithms, is it?

[00:30:48]Right? You freeze them if you're dealing with an offline offline model, ideally.

[00:30:52]Um um and then um in the second part of the course, we we are going to look at um uh some aspect of data warehousing, right? So, we're supposed to discuss uh how we go about designing data warehouses. So, look at uh this whole notion of extraction, transformation, and loading, right? How it fits in into the overall data mining pipeline, I suppose.

[00:31:15]Um and then at at some point, and I think this uh um this learning outcome here is associated with the data mining part here, where we are evaluating data mining models, right? So, once we implement these data mining models, we want to to be able to um understand exactly how we go about assessing the relative effectiveness, is it? Yeah? Uh or their relative success.

[00:31:42]So, we'll discuss evaluation metrics, key evaluation metrics, right?

[00:31:47]Um and then uh they're supposed to be uh a discussion of how exactly we go about applying data mining and warehousing tools. Now, you notice in the program document, which I will put up on the Moodle, by the way, there's there's mention of um there's mention of uh a whole range of of tools, right? Things like Python, Weka, and RapidMiner, or something. In In 472, we are going to be biased towards Python. So, we're not going to look at Weka. Uh Weka is just um it's uh it's a user-friendly tool, right?

[00:32:22]GUI-based tool that makes it a lot easier, I guess, for you to to implement some of these models, ideally. Um so, we're not really going to look at Weka or RapidMiner. We're going to do everything in Python.

[00:32:34]Um and then we're supposed to also uh have very extensive discussion aligned with ethical issues as they relate to privacy, you know, and bias in data mining.

[00:32:46]Very, very interesting topic.

[00:32:48]Uh one that is often neglected here. Um and I I usually take advantage of discussion uh aligned with ethical issues to to extensively look at uh our data protection act of 2021.

[00:33:04]Uh so really really really really excited about about that.

[00:33:08]Um So, 4792 is a half course, right?

[00:33:14]Um of course, you know that, right? You're supposed to be leaving wounds I hear hopefully by end of the year here. So, 4792 is a half course.

[00:33:22]Um uh and uh this thing is going to be run I want to say I I'm saying blended here because I'm I'm being optimistic here. I'm hoping that uh this sanitation issue is going to be sorted out as soon as possible.

[00:33:40]At at least uh uh we don't want the situation where we cover more than half of the content and wait is still going on. I don't know. I'm hopeful though. So, I want to say we're going to run this using a blended learning approach, but also um what I want to do uh uh at the time when we're having these visual interactions is at some point, maybe beginning next week or something, I don't know.

[00:34:04]Uh perhaps the other week. I want us to to start experimenting with what's called uh a flipped classroom approach, right?

[00:34:12]Um I don't know if you've you've you've had uh lecturers kind of like teach using a flipped classroom approach here, but there's there's really a lot that has been written about uh a flipped classroom approach. Incidentally, it's um it's quite effective, right?

[00:34:27]Um so, the idea is simple. We try and see if we can create bite-size uh teaching materials, videos, right? So, uh I would probably pre-record a a lecture session like like I've done right like we're doing right now. I pre-record it, I make it available to you.

[00:34:42]You consume it at your own time, and then we use a session like this to to have a discussion about the content itself. And maybe focus on the um the aspects of the the content that were difficult for you to understand, right? So, we're flipping things ideally, is it?

[00:35:00]Yeah? Um instead of you understanding the problematic things on your own, reading them after class, we're we're doing this in class or something. I don't know.

[00:35:09]We'll try and see if we can make that work, yeah.

[00:35:11]Um and uh ideally, we are supposed to have uh uh 3 hours of of classes every >> [clears throat] >> um every week. Um if this was being done in person, we would have um 1-hour-long sessions, right?

[00:35:31]On three separate days ideally in each given week, but unfortunately, because of our current circumstances, uh things have been uh have been changed slightly.

[00:35:43]And so, what will happen is um we shall have uh um a 1-hour-long session on uh on Mondays and the 2-hour-long session I think on Tuesdays or something like that, right?

[00:35:54]Anyway, but but so, we are we are supposed to have 3 hours' worth of lecture sessions per week.

[00:36:00]Okay? And then there's supposed to be 2-hour-long sessions, tutorial sessions.

[00:36:05]Um have [clears throat] to follow up. Uh we're still We are following up with the head of department trying to find out if uh tutors have been co-opted, but you're supposed to be entitled to tutorial sessions.

[00:36:17]Um if this proves to be a problem like it did in uh 4505 for those of you that were enrolled in 4505, what what I will attempt to do is maybe convert the Monday session into a tutorial sessions and then the 12 hour session will be an intensive lecture session ideally, but we'll try and see what will will happen. Yeah.

[00:36:38]Yeah, so um I think on the Moodle I have made available the calendar, right? You want to bookmark that.

[00:36:46]Um because um things change, right? It's possible that we may kind of like shift things around here.

[00:36:55]But as things stand, um we shall have classes on Mondays between 8:00 and 9:00 and on Thursdays between 11:00 and 13:00. Okay? You want to take note of that.

[00:37:05]Um and you want to make sure that you bookmark the calendar. There's a Google calendar for this course.

[00:37:11]Okay.

[00:37:12]Um All course materials are going to be made available on the Moodle.

[00:37:18]I already sent details um I think in the in the in the fourth year WhatsApp group.

[00:37:24]I I must have included the the URL to the Moodle course site. But then again, you're automatically enrolled into this course, so you probably already have access to this. I guess. I don't know.

[00:37:34]Okay.

[00:37:35]Um So all all course materials, uh assignments, and all of those things are going to be made available on the Moodle, right? You want to make sure you have access to Moodle. If you have uh problems accessing the Moodle and uh because of what's going on on campus, you you don't have any luck, you can write to me and then I can forward your query to the relevant people trying to see if they can help or something like that.

[00:37:59]Um in the course outline that I will make available, you'll notice that there's a series of prescribed and recommended texts um textbooks.

[00:38:09]Um No, no, there's no WhatsApp for this course. I meant the general fourth fourth year group. Uh thank you very much.

[00:38:19]Um there's no fourth year WhatsApp group.

[00:38:23]So, there's [clears throat] prescribed and recommended textbooks, right?

[00:38:26]Um I don't know, right? I mean, the orders change very when it comes to computing, uh what you begin to notice is that uh often there's there's actually better material online um than these often stale textbooks, right?

[00:38:44]Um, but but nonetheless, I mean, these are the things that are in the program document. And then, I think for some of these textbooks, I have personally used them. Certainly, I haven't used data mining concepts and techniques by Han and Kamber, but there you go.

[00:38:56]All the books.

[00:38:57]Um I've not uh I've certainly not used these data mining techniques again by uh This is Linof and Berry. Uh, don't know about that, but uh yet another prescribed {slash} recommended textbook as outlined in the program document, right? Um, what what I just wanted to draw attention to follow these different books is you want to be very careful about uh the editions that you are you're you're reading, is it? So, if there's a a later edition, a more recent edition, you want to make sure that you probably use that book, right? Uh, I mean, 2011, I don't know, right? That's like 15 years ago, right? 1 and 1/2 decade ago. That's a long time here.

[00:39:41]You know, so um uh and also, when when you're using these books, right? When it comes to the technology, you probably want to complement what is being discussed in the book with uh uh what is currently going on, right?

[00:39:57]Because some of the libraries, for instance, that are being mentioned in these books may not really be relevant.

[00:40:02]I mean, look at this, two decades, 20 years, right? So, but there's um there's also data warehousing and mining, right?

[00:40:08]If the encyclopedia of data mining of data data warehousing and mining, uh I've certainly used this book. I remember using this um this is a companion textbook uh for Weka, you know?

[00:40:23]Um you know, so I I I would recommend this, right? The The people be be behind this are from Is it Waikato University or something in New Zealand? Yeah. So, there are many practical machine learning tools and techniques.

[00:40:37]Um in terms of tools, um I mean, this is a hands-on course, right? It's not theory-based. It's It's actually practical course.

[00:40:46]>> [clears throat] >> And um some of the tools that we're going to be using here uh I may have to update this because uh I will not be using uh VirtualBox per se, but I will be using uh I will be using uh uh Python environments. I prefer to use Python environments, right? I mean, I haven't really used uh I haven't I haven't uh created uh um an environment for this yet, but but they're going to be something like this, right?

[00:41:20]Uh so, I'll be using Python environments. I'll create an environment for um for CSC um uh Well, I will share the requirements to TXT file, right? For CSC 4792, so that you can replicate the environment I'm going to be using. And the beauty with replicating my environment is that uh these examples I'm going to be um um uh providing you will only be successful, right? If you replicate my environment ideal. Okay, so so our primary uh uh our primary tool here in our tool set is going to be a Python programming language. Everything is going to be Python-based.

[00:42:02]Um but in the mix are going to be a series of uh libraries, right? That we're going to be using.

[00:42:09]So, you will see me talk about things like pandas, matplotlib, you know, numpy, you know, seaborn um and a few other maybe uh libraries or packages that we're just going to use once off here. But but I think what's important is once we we have a very gentle introduction to to how exactly we go about using these tools, you should be able to um to to probably install and and and use whatever library you you you'd want to use 3D, right? So, things like scikit-learn and stuff are things that we're going to be using in our step.

[00:42:45]Um >> [clears throat] >> Yeah.

[00:42:49]You know, some of these things that are here we may not really use them. Uh for the longest of time I've wanted um to incorporate uh um you know, uh neural networks or something with especially when it comes to discussion of of algorithms or estimators, but we may not really have time to do that, right?

[00:43:08]Especially that that's not really part of the scope of the course.

[00:43:11]Um but important nonetheless, right?

[00:43:13]Because this is a rapidly evolving field here. In fact, some of some of these things that were previously very very hard to do are no longer hard to do, right? With the rise of foundation models.

[00:43:23]Really Yeah, so so um there's a dedicated lecture um I don't know if it's the next lecture.

[00:43:31]I I think so. After the introduction uh I would want us to before we we start looking at the data mining pipeline, I would want us to make sure that we're on the same page as it relates to the technology itself.

[00:43:44]So, we will quickly have an introduction to the the the the the fundamental of the the core libraries, modules that we're going to be using in the course so that we don't get lost when I'm demonstrating these things, right? In subsequent lectures.

[00:44:02]Um um I think I think uh it's a plus that um all of us already know how to program in Python. I'm told Um, maybe now would be a time here. In the If you are able to in the chat, um, I'm interested here. Can you just type yes if you if you know how to program in Python and no if you don't. If you No, yes if you've you've you've um you've enrolled into a Python programming course in the past, right? If if you've done it cuz I don't know if the guys that were doing software uh engineering the guys that are doing software engineering are the ones that did Python. I don't know. Um, if you can manage in the in the incoming messages there, type yes if you enrolled into a course where you were taught how to program in Python and no if not. I'm mostly thinking about the people that are doing systems or something. Woo.

[00:44:54]Uh This is going to be horrible here.

[00:44:59]Then, what what we may do is cuz I wanted to skip the Python programming.

[00:45:05]We're going to have a very a very um uh We're going to have a very gentle introduction then to Python.

[00:45:14]We'll just probably a few Yeah, I'll have to come through these things later on, okay.

[00:45:21]Um, so thanks thanks a lot for those of you that are typing yes and no and stuff here. Be interesting to analyze this data later on, I think. Um, I can put up a poll or something. So, we're going to look at So, when it comes to Python, right?

[00:45:36]There are a number of kind of libraries that we're going to be using, right? We can't run away from scikit-learn because it turns out that the scikit-learn is aligned with uh if you are paying attention, it's aligned with um uh number three here.

[00:45:52]Where we are saying implement data mining models not algorithms, right?

[00:45:56]Most of these things we're going to pull from um scikit-learn library, right?

[00:46:01]Um, so you will see us uh doing things like import uh SKlearn or something, right?

[00:46:08]Then um, from here we'll be able to uh I don't know why my computer is a bit slow in terms of uh you'll see us do things like this, isn't it?

[00:46:19]Yeah? Anyways, um that's neither here nor there. And then there's Pandas, obviously when you're uh wanting to manipulate data um if you're looking for staging area, there's going to be a an easier way of of handling that data, right?

[00:46:34]Manipulating that data, and Pandas um is a go-to, you know, uh library if you will.

[00:46:40]Um so you again you see us do a lot of import, you know, Pandas and then Pandas.read or something to read data, and then from there we can, you know, splice and dice the data or something, right?

[00:46:52]Um, manipulate it in all sorts of different forms, yeah?

[00:46:57]Filter, you know, select desired attributes um or values, if you will.

[00:47:04]Okay? And then um um we Oh, this is interesting. I see there's there's actually 53 people that uh in the uh YouTube here.

[00:47:20]And they're actually people that are responding. Yeah, for those of you that are uh online, please, the question that I asked those those of you that are on on YouTube live um please uh if you can manage, just type no if you know how to program in in in Python, and yes if you do, if you've done a course, I guess.

[00:47:39]Um and then what [clears throat] you begin to notice is that uh you know, if you look at your typical data mining pipeline um there's a lot that happens there, right?

[00:47:49]When you're working with massive uh amounts of data um it becomes important uh for you to visualize that information.

[00:47:58]In fact, because uh you know, it it's usually very very difficult and cumbersome for you to uh to analyze individual observations, right?

[00:48:08]So, what you do is you graph, you summarize information somehow, compact it, right?

[00:48:12]And it turns out that there are dedicated libraries that can be used for that.

[00:48:17]Um and well, one of the most widely used libraries >> [clears throat] >> when it comes to Python is actually Matplotlib, right? So, you will see us do this as well, right? You see us do I can port Matplotlib.

[00:48:31]Or from Matplotlib import pyplot or something like that, right? So, Matplotlib import uh pyplot, right? As PLT or something like that, right? And then you know, uh you start graphing and stuff, right? So, so we'll again look at exactly how to use, you know, uh libraries Matplotlib.

[00:48:49]And then, um uh you know, when it comes to data mining, there's something interesting about data mining.

[00:48:55]The goal is usually to build these models, right?

[00:48:57]Yeah?

[00:48:59]And the interesting thing about models really is that um they they're not really user-friendly, right? In their raw form, if you will.

[00:49:11]Okay?

[00:49:12]Um and I want to give you an example of uh I'll give you an example of uh or Deep Seek or something like that, right?

[00:49:20]It's like Deep Seek or something like that. Well, I'm uh Deep Seek or something like that, right?

[00:49:27]Or the llama uh or or yeah, Deep Seek or or Meta AI or something like that. Well, it turns out that behind behind Deep Seek, right? Be- behind that Deep Seek chatbot, when you go to chatbot.deepseek.com, right? What you see is this nice interface where you can just type in text, right? In that text box. Yeah, you key in your prompt. Uh I mean, of course, there are other configuration details that you you you get to change. Um but it turns out that behind the scenes is what's called a model, is it?

[00:49:58]Right?

[00:49:59]Um and So so there's a model, right? And typically because the model is not very user-friendly, unless if you're a technical person, you won't be able to even be able to use it. I mean, I can use it. I can download DeepSee model, right? R3 or something like that.

[00:50:13]And be able to use it on my computer, right? But the average person will not be able to do what I do, what I'm what I'm capable of doing, right?

[00:50:20]Um and so what we do is we build an interface, is it? Now, it turns out that when you're wanting to build a model, you follow what's called a pipeline.

[00:50:29]Um and one of the key things about a pipeline is you want to build it in such a way that it's reproducible, right?

[00:50:36]That somebody else can be able to replicate your steps.

[00:50:39]Um and as luck would have it when it comes to data mining, for instance, uh or machine learning or AI, it's usually advisable for you to uh to create this pipeline using what's called a notebook.

[00:50:51]Right?

[00:50:52]Um So a notebook, as the name suggests, is precisely that, right? If you remember your first year, you when you're doing those experiments in chemistry and physics, if you did that, um I don't know if you were you were taught to do this, but you'd be taught that you needed a lab book or something. I remember when I was first year myself, 2000 and um 2000 and uh and three it was, we literally had lab books, right? Where you would write down things, procedures, and all those things, and results. Well, it turns out that a notebook is something similar, right? Um and and this note this Jupyter notebook is is nothing more than a web service, a web application, actually.

[00:51:32]Right? So within that web application, you you include interesting things like the raw code itself, this Pythonic code, textual content, and you know, visualizations or um I guess it would be like equations, right? Mathematical equations, if you will. Um difficult when there are a lot of people to mute ask you to mute or something.

[00:51:52]Yeah, so a whole range of tools, you know, um I don't think we're going to get to a stage where we're using, you know, frameworks like Keras or PyTorch.

[00:52:00]Um but if time permits, maybe we can we can we can look at these as well.

[00:52:06]Now, the course grading here is a bit a bit tricky, right? It's a bit tricky because of what is currently happening.

[00:52:14]Um but as things stand, we're going to we're we're going to follow this this breakdown here. It's going to be 5%, which is going to be tied to participation and engagement. And traditionally, the way that I gauge participation is by taking attendance, right? So, what I will do is >> [snorts] >> Uh so, if you're on on YouTube, by the way, just uh I'll ask you to make sure that you type your your name and your email address or something like that, right? So, just your name and and your email address.

[00:52:52]And I'll check your username, by the way, and your email address like this, right?

[00:52:56]Um yeah, and just paste it in the chat on YouTube. Um the people that are currently online, um it's very easy for me to kind of take attendance. It's um can automate this process. It's not It's not a big deal, right? Um So, anyway, so 5% is supposed to go towards participation and engagement. Now, I'm saying it's a bit tricky here because I'm well aware of the fact that uh because um we are currently not on campus, you know, internet connectivity may may prove to be problematic.

[00:53:25]Um you know, so we'll try and figure out something. Um maybe >> [clears throat] >> if if I have a bit of time here, I'll probably create activities on the Moodle to try and assess whether or not you're engaging with the course. It will be better than taking attendance or something, you know. There's also supposed to be labs, there's supposed to be quizzes actually. Um and obviously, I mean, there's there's nothing like doing online quizzes because everybody is going to pass, right? Cheat and pass, is it?

[00:53:54]So, we'll probably defer these up until we get back to campus. Usually, these are these are converted into quizzes, right? So, quizzes on various topics that we have covered ideally. And then, we're supposed to have um we're supposed to have uh um a practical. So, it's a hands-on practical project.

[00:54:15]And my approach when it comes to the hands-on practical project is uh I have you guys self-organized into groups, right? You self-organize into groups. Um for now, I assume it's like five or something like that. You self-organize into groups and then at a later stage, when you self-organize into those groups, you will choose a topic that you want to work on. I will prepare topics.

[00:54:42]Um and then you work on this project.

[00:54:44]So, so the idea behind the project is you work on it uh beginning now all the way up to at the end of the course, which is somewhere in August or something. And in between, there'll be like checkpoints, right? I expect you to uh submit deliverables associated with uh various components or aspects of data mining and warehousing.

[00:55:04]Um uh So, so yeah. Um and it's the internal an assignment has already been been circulated. So, assignment number one is available on the Moodle. I've also sent details by uh email. Uh I'll talk more about about that towards the end here.

[00:55:21]Um and then, um there's supposed to be um obviously um class theory test, is it? Now, it's possible that uh the class theory test will probably be allocated slightly more marks here because of what is currently happening, right?

[00:55:37]Meaning that we will wait until we are back on campus before we can write this class rate test because it has to be in person, ideally.

[00:55:44]So, you notice total here is about 40, right? The the the course here is designed in such a way that 40% is allocated to to the continuous assessment score and the 60% is is [clears throat] um is allocated to the final examination.

[00:56:02]Yeah? Um I just to kind of give you an idea of the breakdown here. This is spreadsheet from last year or something.

[00:56:11]You know, you want to take these things seriously, right? You're in fourth year.

[00:56:15]Uh take this seriously.

[00:56:18]Um if you're told that marks are distributed as follows, right? You want to be serious about this.

[00:56:26]Um I I don't want a situation where you're putting me on the spot, right? You're writing very very strange uh letters. I don't know if it's to the dean or something, to the VC, requesting for a grade to be upgraded when you were not engaging with the course, right? We check we check participation and you never attended not even a single lecture.

[00:56:51]Right?

[00:56:52]We we check um your your score for the group and you never your group never submitted deliverables, right? And you admit to it.

[00:57:02]You're you're shooting yourself in the foot here, right? Um at fourth year, I mean, if I were you, my goal would be I want to move on, right?

[00:57:10]And the only way you'll be able to move on, sadly, is if you're serious with work, right? So, that you clear the courses or this course at least.

[00:57:17]Anyways, um so, just showcasing how how the the breakdown of the marks was done at the end of it all, right?

[00:57:25]So, what you notice here is uh >> [snorts] >> you know, it's class participation here, right? Which is just attendance. Sad, right? You had people that were literally neglected um participation. You tell them at the beginning of class like I'm telling you right now, right?

[00:57:41]To say we're going to take a we're going to figure out a way of taking attendance and um figuring out if you're engaging with the course and you don't engage with the course. Giving up five marks or something like that, right?

[00:57:52]But anyways, and then there's the labs.

[00:57:54]These were quizzes. Uh we had uh we had four of them, right? So, they were out of 2.5 obviously.

[00:58:01]Um and then we had class theory tests, right? Besides that, in the middle here, you you may not really see this, but there's uh there's a group component here, right?

[00:58:09]So, we had uh there's a report, there were presentations, you know, [snorts] we we we had people demonstrate things.

[00:58:17]We had people submit notebooks and GitHub repositories where we were checking the commit messages, right? And the checkpoints, right?

[00:58:28]Uh so, all of these would be would be provided to you in the form of assignments, right? Group-based assignments.

[00:58:35]Prepare an assignment, we make it available on the Moodle, and you work towards it at the end.

[00:58:40]Um I think we're already we're already aware of the grading thresholds associated with uh undergraduate studies at the UoN Science and the School of Natural Sciences or at the UoN already.

[00:58:49]So, there you go, right? Um if I were you, I'd be striving not to to fall in the D+ D+ D range and F range here, right? And and really, if you just engage with the course, I promise you, it's not a difficult course, this one.

[00:59:05]You will pass, right? It's it's um it's practical, it's something that you can easily relate to, and it's pretty easy, yeah. Uh in terms of course instructors, uh this year I'm the only course instructor, right? So, you're stuck with me until August.

[00:59:20]Sorry if that's a disappointment. Um something interesting about office hours here, when I'm [clears throat] when I'm on campus, I normally have office hours when people that need to come and see me or anything related to either the course or your stay at UNSW or anything else that you think I may be of assistance um can come and see me, but because of what's what's happening we were actually also told to stay away from campus cuz there's literally no water, right?

[00:59:47]Um so so I've I've been working from home for the last week or two or something we can have I guess.

[00:59:53]Um So my office hours my virtual office hours are going to be Mondays from 9:00 to 10:00, okay? Take note beginning next week. So if there's something that you want to see me about what I will be doing is once we end class, I will stick around.

[01:00:08]I will be available I'll be online. I'll be in in the call and what you do is you can just, you know, log in online and then just uh chat to me about about whatever it is you want to chat about or something, whatever problems you may be experiencing or something.

[01:00:22]Um the link we're going to be using for these interactions is the same link we're using we're going to be using for classes which is that, right?

[01:00:29]Um if you if you're unavailable during this period, uh you can obviously, you know, reach out to me via email um and stuff.

[01:00:37]Okay?

[01:00:38]If if that makes sense uh or something like that.

[01:00:42]Actually glad there's uh something interesting here on YouTube live there's about 45 people, right? And uh here it's quite quite a bit here about closely 100 which is good stuff. It's working this YouTube streaming is working. And then in terms of communication so our communication in the course there's no WhatsApp here.

[01:01:01]It's exclusively via email and you must understand this, right? Okay, maybe >> [clears throat] >> maybe I'm being resistant here, but um I'm a huge detractor of WhatsApp. I hate WhatsApp, right? For future communication. It's difficult to keep track of what is happening.

[01:01:17]You know, there's a lot of noise usually in WhatsApp groups.

[01:01:20]So um communication is exclusively done by email. If you have an issue and you want to reach out to me, send me email. There we go, that's my email address right there.

[01:01:30]Right? If you have a concern, right? Use a mailing list, [email protected].

[01:01:37]Boom, right? This is the mailing list here.

[01:01:40]So, all communication will be done exclusively. And you must understand, right? I was counting how many people are you noting here? I was counting I think there's uh uh I I counted I think close to about uh 200 or so people something, right? In this this course.

[01:01:56]Right? So, if you're communicating to me via WhatsApp here, it's it's a bit of a problem, right? It's going to appear like uh I was literally counting here. It's going to appear like uh It's going to appear like I'm not Yeah, 225 people. So, yeah, twiddling my thumbs here.

[01:02:15]Um interacting to this number of people is just not going to work for me. It doesn't work for me, right?

[01:02:21]So, there you go. Uh communication is going to be done uh exclusively via um via email, via the mailing list. You want to make sure that you you you you you're actually subscribed to the mailing list. I sent details about this uh if you will.

[01:02:40]Um Listen, um I don't like uh having this discussion because especially when I'm dealing with senior students, right? You've been at UNZA for what? Since 4 years now and counting.

[01:02:54]Um you know about academic dishonesty.

[01:02:58]If you If you are caught engaging in any form of academic dishonesty in this course, you get a zero score.

[01:03:05]Right? Uh don't force me to do that.

[01:03:08]Uh it's it's very sad what is happening, right?

[01:03:12]It's very sad what is happening at UNZA.

[01:03:14]It was never like this when I was a student myself.

[01:03:17]Literally, I was administering a test to postgraduate students last week, yeah?

[01:03:23]And this person pulls out their phone, right? They're literally copying things from their mobile device.

[01:03:29]Uh you know, that's uh that's not going to fly here, right? Not with me at least, right? Um so, I don't have to sit here and and and and rant about academic dishonesty or care about academic dishonesty. What I'm going to do is I'm going to refer you to the rules and regulations on academic assessments for students, right? It's a document um official document uh that outlines all the different offenses and implications if you're caught doing that, right?

[01:03:59]And I'm going to apply these uh in this course religiously.

[01:04:04]Okay? Um um religiously, you don't want to get a zero score in the course because I mean in the assessment because ultimately, I mean, what that means is that you're obviously not going to you're not really going to write the exam, right? You have to pass this year for you to write the exam or something like that.

[01:04:21]Right? There's various forms of academic dishonesty. These will be outlined >> [snorts] >> in the assessments as they're made available, especially when it comes to written assessments and stuff. You know, I um I don't want to act like I'm an enemy of progress here.

[01:04:37]Um um um I'm well aware of what is happening um when it comes to AI, right?

[01:04:46]Especially LLMs, at least you know, chatbots and stuff.

[01:04:50]Um when it comes to implementation you know, um I'm not really very strict, right? Uh but but we'll be specifying ideally. We're going to be specifying >> [snorts] >> exactly what is permissible and what is not permissible as these assessments are made available to you ideally. But just take note here, academic dishonesty will not be tolerated.

[01:05:09]Right? If you are caught cheating, somebody was caught in some course, right?

[01:05:14]What?

[01:05:15]Scratch that. In this course, somebody was caught cheating using ChatGPT in the test. Right? Don't want to do that.

[01:05:23]If you do that, you're going to come back and do redo this course.

[01:05:27]Right? And I don't want to be forced to to do that, right? I don't want to do that. Please don't force me to do something I don't want to do.

[01:05:34]Um something that I forgot to include here um is uh I'll ask you to please have the I'm I'm I'm asking you kindly to nominate two course representatives, one male, one female, and have them reach out to me and tell me to say they are the course reps or something. They can WhatsApp me, the course reps alone, or they can email me.

[01:05:56]Uh it's important that we have course reps so that there's a bit of there's some semblance of order at least, right? I mean, so this chaos.

[01:06:03]Um and then just to mention here that uh assignment number one has been made available on the Moodle. I've sent email about this. It's a very simple assignment.

[01:06:13]Your role your task Okay, your role your task is very very simple, right?

[01:06:19]Between now and Friday, find four other individuals who you want to work with.

[01:06:26]Right? Remember I mentioned this group-based project which accounts for 15 marks? It's part here, right?

[01:06:32]What you're going to do is you're going to find four other people. You You self-organize into groups of four.

[01:06:37]Okay?

[01:06:38]And in the assignment, what you notice on the Moodle is there's a direct link to a spreadsheet.

[01:06:44]Okay?

[01:06:45]And in the spreadsheet, all you do is very simple task, right?

[01:06:49]In the section in the columns under student ID, all five of you must just indicate your student IDs here or here or here, all five of you. That's it.

[01:07:00]Okay? You must do this uh by Friday.

[01:07:03]Friday is the deadline.

[01:07:05]Right? Uh we have a way of tracking uh whether people are going to go beyond Friday here.

[01:07:11]All right. Um I see this is 9:09. I do apologize. I know you don't have anything uh lined up here on the timetable.

[01:07:20]Because um Mrs. Kumar had actually >> [clears throat] >> had actually initially uh uh allocated two sessions on Monday, but unfortunately I have other commitments on my side. So, I asked that we have a session today and two sessions on Thursday. Okay? So, I'm going to pause here and uh find out if if you guys have any questions.

[01:07:47]If you are on uh if you are on YouTube live, uh you can type your question in um in the chat and and then I'll I'll be able to address it.

[01:07:58]Uh if you are if you're on Google Meet, um you can just unmute your microphone and then uh ask away, right?

[01:08:10]Any thoughts or questions about this?

[01:08:11]Complaints?

[01:08:21]That's weird. Nobody has questions.

[01:08:24]What is going on here?

[01:08:31]Good name five.

[01:08:33]Chat rate I don't know what chat rate is here.

[01:08:36]But you guys have no questions?

[01:08:41]You really thought uh you thought um class was going to be And I know you guys are crazy, right?

[01:08:47]You students are crazy. It's always like uh most of you were happy that your union was like, "No, we are against online learning." I'm sitting there and I'm thinking, "Wow, fourth years that should be excited at the fact that uh school must continue so that you quickly start working, right?

[01:09:03]Uh what of situations where we need rapid response? What mode of communication is that? Well, if you need rapid response, uh course reps usually course course reps have access to my mobile number. I mean, you all have access to my mobile number, right? But what I do is I ignore uh Well, I will pick up calls from students, but the moment they tell me I'm a student in 4792, I'll just tell them, "Listen, use email, right? We need a traceable record here." But if something urgent that affects all the students, right? This is why we need two course reps.

[01:09:35]Course reps will have direct access to me, and they can call me, they can SMS me, they can WhatsApp me.

[01:09:41]We actually usually have like a dedicated uh WhatsApp group with the tutors, um the course reps, and myself or something, the course instructors.

[01:09:51]I don't know what Dhaka Prince is all about there, yeah. And I see nobody has bothered to load the assignment here. Guys, um are there any other questions about this?

[01:10:07]Yeah, I'm going to reach out to you.

[01:10:09]>> Uh yes, I have a question.

[01:10:10]>> yes. Yes.

[01:10:15]Uh attendance will only be recorded on YouTube Live, not here as well.

[01:10:20]>> no, here as well.

[01:10:24]Okay, so only uh the name and school email, not the student ID. No, no, no. So so for for you if you're online here on Meet, that's fine, right? I I already have uh attendance. I mean, I I already know who is attending right now, right?

[01:10:39]Um not difficult here. But but YouTube Live is a bit difficult for me because uh I can only do this retrospectively. Is it? I'll have to I'll have to go to uh I don't know if I'll have access to all the uh Well, for YouTube Live, I can only know if somebody has attended if they type something in the chat. So, I'll go and harvest the the these 52 people that are in the chat here.

[01:11:05]Um I'll go and harvest their details.

[01:11:07]And as we go on here and we'll fix this. I'm trying to see if we can fix the issue of the 100 uh participants limit with Google Meet.

[01:11:15]It says it's it's taking longer than usual. So, it says it should By next week, we should probably address this issue and then we'll be back to normal here. But just in case uh is that fine?

[01:11:26]Oh, all right. Yeah. Yeah.

[01:11:29]Any other thoughts or questions about What What don't We don't have access to the spreadsheet. Well, you must um What you must do is you must request for edit access, right? Uh The reason I I ask people to request for edit access is because uh in the >> [clears throat] >> you know, what people do sometimes is they do funny things, you see? They'll do funny things. Or what I'll do here is I'll just add the mailing list then.

[01:11:55]There we go. So, everybody who is If you're already on the mailing list, you now have access, edit access. You should have a edit access. If you don't, uh open the spreadsheet and then request for edit access.

[01:12:10]Request for edit access.

[01:12:16]Okay? Um any other thoughts especially as it relates to um what do you call this?

[01:12:24]This administrative here, right? I want to make sure that uh we are all on the same page.

[01:12:31]Uh I have got a question, sir, but this is away from this course. Oh, yes.

[01:12:36]Yes, uh this is pertaining to 4505. Uh we don't know if the CA has been updated or not.

[01:12:43]Uh yeah, I've done I'm I'm I'm done with the marking. You know, I when they canceled this thing, I I got uh a bit lazy in terms of updating the CA. But I will I will add them to the spreadsheet and then you should have access to them.

[01:12:56]And then uh No, the test results the test the actual test scripts you can pick them up once we are back in person.

[01:13:03]Uh I don't know if uh Miss Miss Deborah and Mr. Lubasi are somewhere closer to campus. They can reach out to me if they want to fetch the scripts, but I would much rather um you get them from me once we are back on campus.

[01:13:19]All right, sir.

[01:13:22]Okay, guys. This is 9:15. Uh if there are no further questions, um and and I see am I the only one experiencing I don't know.

[01:13:33]Sorry sir sorry sir another question.

[01:13:35]Yeah.

[01:13:36]Yes, about editing the spreadsheet. Is it one person from the group or it's everyone? No, no, no. I I mean it would be nice if everybody can do this themselves, right? If you want you can do it for your colleagues. Well, but why not?

[01:13:49]Why not just have everybody do this, right?

[01:13:52]Everybody should do it.

[01:13:55]You understand? Or if you want you can designate somebody to to enter the cuz you're just entering computer numbers here, by the way. Just here.

[01:14:02]If you want you can choose one person who is going to to add uh to add uh names for other people, right?

[01:14:09]That's fine.

[01:14:11]10% air and 90% my work. How's that? I don't know what that is. I don't know what Humphrey Kinyua is uh is trying to say here in in the YouTube thing, yeah.

[01:14:22]Um When we When will we get access to Google Sheets? It's not opening. You want to request uh So, NOCs is online is is on YouTube live and he's trying to find out when they'll get access to the spreadsheet. You should have access to the spreadsheet. I see there are already people that have access to the spreadsheet. Who is this?

[01:14:41]This is Mwansa has access.

[01:14:43]Um Whatever that means, yeah. Um so, just make sure when you get access to the spreadsheet, request for edit access. I can't I don't want to make it open because it'll be difficult to keep track of who is making the changes, right? I want to I want to to to see exactly who is doing what.

[01:15:02]Just in case there are funny things that people In the past, right? The next stage is you will choose a topic, yeah?

[01:15:08]Topics have not yet The next assignment you choose a topic. And last year we had people that were you know, editing out things here, right?

[01:15:15]So, we're trying to avoid that.

[01:15:17]You're deleting entries for your colleagues and and stuff.

[01:15:26]The tools So, there's a question here from from Zake, right? He says the tool on on on kind of like this thing here.

[01:15:36]YouTube live. He says the tools we are getting, can they work on a low-end PC?

[01:15:40]So, if you look at the primary tool you're you're going to be using here is in Python, is Colab, right? Google Colab.

[01:15:48]>> [clears throat] >> And it turns out Google Colab is I think it's designed as a progressive web application, right?

[01:15:55]If you look at it. So, if you go to colab.research.google.com the primary tool we're going to be using is Google Colab. So, if you go to colab.research.google.com colab.research.google.com on your mobile device, on your mobile device, you'll notice that it's a progressive web application. And literally, you'll be able to run these notebooks on mobile device.

[01:16:16]Right? So, so there we go and Zake or something.

[01:16:23]Okay. Any any other Any other questions?

[01:16:28]Okay, if not, guys, thank you very much.

[01:16:32]I know this is going to be a bit hard, right? Transitioning.

[01:16:37]But look at the bright side, especially for those of us those of you from from our department who are doing computing you should embrace this.

[01:16:45]It's a good thing. If you have suggestions on if you have complaints about the user experience or something it's important that you reach out to me.

[01:16:53]You have access to my email address. You have access to the course reps so that we address these things, right? If there's something that is making you I don't want to use the word uncomfortable here but that is hindering your ability to comprehend or understand the things that we're going to be looking at to us let us know early on in the process so that we try as much as possible um to to make this experience uh somewhat more bearable, right?

[01:17:19]A good experience. It should be a good experience. I'll try as much as possible to record these sessions so like I'm recording this I'm streaming it.

[01:17:26]Uh I mean if I'm streaming it I'm just going to add it to the I'm going to add this stream uh to the CSC 4792 playlist.

[01:17:36]Um and I'll share the direct link um on Moodle, right? So there's a playlist that I'm going to create create which will have all the live streamed classes and pre-recorded