- 29 reviews
- 28 completed
The best cryptography course ever. You get everything you usually don't get at a cryptography class from one of the most recognized researchers in the area. The course is a roller-coaster ride to the world of cryptographic primitives, different attack methods and how to prevent them. The quizzes are difficult and challenging,the optional programming assigments are also difficult and very interesting. It's quite an honor to be able to take this course for free. A total blast!
This course has good content but the way the content is presented is just terrible. The instructor doesn't care about making the concepts simple he just mumbles around formulas and papers without examples. As any student and instructor knows complex subjects can be explained if you create simple examples to show how things work in practice this course doesn't care about that. I also disliked the programming assignment that uses a very strange platform and asks the students to complete very simple tasks without any interesting goal. I liked the topics a lot and I disliked how they were presented even more.
Where is a sixth star when I need it? This is just a fantastic course. The lectures are perfect, clear simple, well explained and with plenty of examples. The programming assignments are fun to do and challenging. The instructor is all over the forums helping, fixing bugs, coding along with the students, answering anything. I've never seen that level of commitment in any other course. Hats off Mr Sriram. Anyone interested in Linear and Integer Programming will find the course fascinating. Totally recommended.
This is a very nice intermediate/advanced course on Social Networks. It discusses several topics that are usually not mentioned in other courses and the instructor is one of the most important figures in the area. There are some examples of applications of the topics seen. I would have liked some programming assignment or hands-on application to use the concepts and idea for some specific goal to make them least abstract. This is highly recommended to anyone interested in Social Networks and networks or graph theory in general.
A very nice introduction to neworks including social networks, models and economic networks with some game theory. Prof Kearns is really terrific explaining, clear and precise. His only fault is his total lack of knowledge in the field of geography :) There are well designed quizzes to practice the content of each video lecture no programming, exams or any other assignments and the whole material is available from day 1 so you can really go at your own pace. I'd like to say something about the comments mentioning prof Kearns reading long slides. He is not reading, he uses the slides to highlight the topics he explains, the slides are very well constructed because they can be used as a summary of what was discussed in each lesson. A really beautiful course with a great instructor, not difficult at all. Highly recommended.
The most useful thing I can say is: This course is NOT for computer science or programmers. This is oriented towards Information Science people, archivists, librarians, researchers, etc. The course covers different metadata formats, thesauri and standards. The instructor is very good explaining the topics but when dealing with CS stuff he makes many mistakes some of them really very big. There are easy quizzes after each week avalanche of videos, the instructor is quite slow explaining and takes a lot of time even for very simple concepts so I recommend watching the videos at least at 1.5x speed. If you are looking for a good chat about metadata vocabularies in information science this is a good course. If you want to learn how to use and process metadata this is not even a start. Edit: I've realized this course has just too many critical errors to be recommended, if you are interested in the semantic web and metadata look elsewhere because if you take this course you will need to unlearn everything and then learn it the right way again.
This course is about writing web applications using R and Shiny. You are not going to learn anything here that you can't learn on the web if you need to use shiny. Just follow the shiny tutorial and you are good to go for the first quiz and the project. And that if is the big problem in this course. As a part of a specialization you are forced to learn a tool that has nothing to do with data science. The lectures are chaotic, jumping from topic to topic without a clear explanation of anything. After a few lessons the instructor uses cartoon animations which would be fine if the course would be aimed to children. These animations only make this course really sad, the time taken to produce the animations could have been used to create a good lecture and good examples about how to use shiny, how to do different things and show more examples. Shiny itself is quite a horribly done thing, the syntax is chaotic the way things are coded are very dirty and the results look very bad. For web development there are really much better tools. A bad course for a bad tool that I hope you don't ever need.
This course has changed a lot it is now much better fully deserves half a star. In the new version of this class the videos are much improved, now there's a guitar in the background that randomly appears and disappears as the lecturer mumbles over formulas without really telling you what is going on. There's also a "project" where you have to do a simulation and then some work that you really never learned in the lectures. You need somebody gifted to explain statistic topics to a general audience and make them really understand the theory, apply it to the practice and make it fun. This is not the case. The examples include diseases, tooth grow, height of children and more diseases. You have to work hard to find examples less attractive than those for a class.
An excellent introduction to the world of machine learning. The way the material is presented is fanastic. The lessons are crafted to teach the student how the algorithms work and how you can code them along with the basic mathematical foundations. Other ML courses start with the math and are difficult to follow. Every time Andrew explains a topic you will think "I can do that" and you will in the programming assignments. Andrew Ng is probably the best instructor a course like this can have and his lessons are a treasure. Fantastic course in all aspects.
This is so far the best course from the specialization. And I'm giving it 1 star so you can imagine how the previous courses are. This is about writing reports in RStudio using Markdown and RCode. You just need to learn basic Markdown and how to insert code chunks to create your report. That takes about 26 seconds. Once you do that you can answer the quizzes and do the assignments. Both peer assignments are just about writing a report doing some basic stuff on a given dataset. The topics are incredibly boring, a hallmark of the specialization. The quizzes are trivial, the assigments can take some time processing the data, creating the plots you are asked to display etc. I only wish the topics could be at least interesting or entertaining but no. The videos might be skipped, there's a lot of material just to fill 4 weeks and justify yet another course.
This course is a scam, as well as most of its reviews here. The reviews were written before the course started with the intention to publicize the course. Please note all the five star reviews about how the course WILL be. Are pre- reviews allowed here? The whole idea of the course is to sell a book and big- data is a good way to catch customers. At least the Foreman grill could actualy do something! A shame!
This course is a gem so I urge everyone interested in these topics to take it. Before explaining why I will start with a disclaimer. This course is one of the first in the category "abandonmoocs" you have the lectures and the quizzes but you are completely on your own, the instructor is not present and there are no TAs or any kind of help whatsoever, so in some way this is the same as just a list of videos on youtube. But the content is terrific and the instructor is a genius!. This course covers very very interesting topics such as PCA, Independent component analysis, image segmentation, signal processing and many other topics and the way the topics are presented is terrific. First you get a great lecture about the theory behind each subject, the instructor takes time to make very difficult concepts simple and he does it without losing depth in the topics. Not only that but he is also fun and presents really nice examples for each topic. Then you get a "practice" lecture where the instructor codes the different algorithms in real time using Matlab, this was awersome as you could just reproduce the code at the same time and understand how things work and what very nice things you can do with the algorithms. There's nothing remotely similar to this course so I give it six stars even in its current state of abandonment.
This course was a major disappointment for me. Everything was bad: content, presentation, evaluation, programming assignments. You name it. I will start with the content of the course. Text Mining is a very exciting area of computer science getting a lot of attention these days, for some reason the algorithms and ideas that are being used today and are the state of the art weren't covered in this course. There's nothing about things such as deep learning, wordvec, glove, recurrent neural networks or LSTM that are the most important tools used today for natural language processing. So I think the content is outdated and I felt I was getting a review of things the instructor likes instead of things really used in the field. Next the presentation: really terrible. The instructor dances around formulas and papers without a single working example to check how things work or any attempt to simplify the topics or explain the most important concepts, just references to papers and reading out loud formulas. The evaluation is also problematic you have only 1 chance at each quiz so effectively you have four exams. This wouldn't be a problem if the lectures contained anything remotely similar to what the quizzes ask. If you are going to give students one attempt at each quiz you need to at least do some examples of excercises in your lectures. The programming assignment was the cherry on the cake, confusing, badly prepared and didn't make any sense. The META system used is really terrible to work with and is a tool that as far as I know is not used at all in the real world. There are plenty of interesting things and tools to use in this area but for some reason META was chosen. I think that was a very bad choice. At the end I completed all quizzes and passsed the course and felt I learned almost nothing and the very few things I remember about the lectures are things that I know won't be useful in the real world. I would have given it 0 stars but I have no such an option.
This is a good course on Clustering algorithms. The instructor covers a lot of topics and provides good examples to understand the most important algorithms. I liked the detailed and numerous examples that were explained through the course. It is not a perfect course as real-life applications of clustering are missing and the programming assignment doesn't make any sense but overall I liked the content and how it was presented.
This course has some nice content but the presentation is very boring. The instructor does take a good time and effort to provide examples of the principal topics and algorithms. By taking this course you will learn several interesting things in a good degree of depth. A good course considering it takes just 4 weeks.
I just finished the last week of this course and I find very difficult to write a review about this course, it has a lot of positive and negative things at the same time. I'll try to sum up the pluses and minuses I found: Positive things: \- Extremely well prepared material, lots of examples, clear explanations, top-quality video and audio. Top-notch. \- Fantastic instructor. Clear, precise and always right, the concepts are delivered in a perfect way. \- Excellent response time in the forums and support to the students. \- Very well prepared quizzes with many questions to make you learn the materials and apply the concepts to the questions. Negative things: \- The course is too long, too many topics, too many videos, it could have been split in 2, 3 or even 4 smaller courses. \- Peer Assignment: it is too long, the dataset is totally boring and uninteresting and it only counts for the distinction certificate. \- The tools being used: Prom and Disco. Disco is a commercial tool and Prom has many flaws and bugs. \- Lectures are too long, too many videos per week and the pace is too slow. (It is ok at 1.5x video speed) I started this course with a lot of energy and liked its start very much, my interest quickly waned to the point of wanting to drop the course badly. The course fails to deliver a real world case to apply the tools being used for something interesting, something fun, something worth it. The peer assignment was a huge letdown as I was expecting a fun application of the topics learned and instead found a longer tool-quiz where you have to use Disco and Prom to process yet another boring and uninteresting dataset. My conclusion is that process-mining is a very interesting field where very cool algorithms can be applied but in the end the work is boring and tedious and the tools that we have available today are not fun to use and full of bugs. It's a good course so long and so detailed that you will learn to love or hate the topic unfortunately for me it was the second. I can give this course 1 star and 5 stars at the same time. You be the judge.
This is a great course. The topics are modern they are presented in a clear and interesting way along with many examples. Gephi and Netlogo are used to create simulations, models and analysis of social networks. Something to highlight about this course is that all of the examples, excercises and programming assignments really made sense, using interesting data for interesting and diverse problems. I loved the idea of a final peer-graded project with 3 different optional areas to cover and an open subject. It's fun and interesting to do and fun and interesting to grade and see what the other students did. This is a fun course with many interesting topics presented in an excellent way. Totally recommended.
This course is a real gem. The best about it is the content and how it is presented, the topics are modern, the instructors know the last stuff about the topics and the lectures are really excellent. Then there's the usual set of weekly quizzes and a final exam. I didn't found any organizational quirks or problems in how the topics were presented. I'm really happy content such as this is made available to the general public online the value is incredible. Highly recommended to anyone interested in modern algorithms and topics about large datasets.
This course on game theory has two fundamental flaws in my opinion: The first problem are the video lectures. The lectures in this course are not long enough or detailed enough to understand the concepts. They do cover the theory but not enough practical examples to help the students take the problem sets. The second problem are the examples. Only a few examples are given and they are completely uninteresting. Battle of the sexes, prisoners dilemma, matching pennies, none of those games make any sense or have any real practical application. I would have loved a peer reviewed assignment were the game theory concepts could be applied to a practical real world problem. I leave the course with the feeling game theory is only a theoretical concept, mostly based on common sense without a real application to the real world.
This is the best, and probably only, online course about automata theory, Turing machines, languages and complexity. From the most basic deterministic automatas to Turing machines and P,NP problems everything is covered with rigor and accuracy. Jeff Ullman is the ultimate authority in the subject and it is a gift to be able to see his lectures. The quizzes are of medium to hard difficulty and there's a final exam so it's not an easy course it makes you study and learn.
A very difficult course with some major problems. The first problem and the reason I dropped is that what you learn in the lectures is not enough to do the homework assignments, and that's totally frustrating. I believe easier exercises with nicer examples that can be done based on what is learned in the lectures would fix this course. The second problem is the boring way the lectures are given without any interesting examples or motivation, just a readout of formulas and theorems without any story between them. Formula 2.3, then 2.4 then theorem 2.5 etc etc. I would have loved this to be a nice class but it fails. Badly.
Eclectic, fascinating, full of mistakes and mysteries. This course is not for the weak. First of all I would like to point out that you need to have experience in Python, SQL and programming in general otherwise this course will be just too difficult to complete. Each week there's a new topic and new assignments, totally different and all of them quite fun. Including sentiment analysis of twitter, matrix multiplication in SQL, Map-Reduce demonstration in Python, Writing and running Pig scripts in the Amazon AWS cluster, a Kaggle competition for machine learning and a visualization using Tableau. The assignments appear randomly, the instructors don't answer questions, the course is full of mistakes, glitches, missing links and many quirks that can be solved via some team work in the forums. It would be a mistake not to give five stars to this course because of those glitches and errors, they are just like life itself. The amount of things the course covers, the fun assignments and the very interesting topics makes this course really wonderful and that's why I'm giving it full marks, you will learn a lot, you will suffer some and you will have a lot of fun.
Machine learning is a very hot topic today. This course aims to teach how to use ML algorithms using the caret package in R. Caret is a wonderful package that is only scratched in the lessons, there are several fantastic tools in caret that are not even mentioned. There's also very little about how to apply machine learning algorithms in real problems or advice about how to use caret for real world projects. The course has 4 quizzes that are easy to solve and then a final project where you are given a dataset and you have to apply ML to predict values of a test set. Unfortunately the logistics for the project are horrible and the way it is graded is really bad. This is as bad as a course can be because it covers a very interesting topic a very interesting tool and yet it fails to be entertaining or even educative. A complete shame.
Regression models is about fitting linear regression models to data. The lectures are bad, with many formulas that are not needed and take a lot of time and space that could be used to show better practical examples. There's a project where you have to apply what you supposedly learned. The project is quite fun to do but you have to research and learn the methods yourself. Regression models is quite a fun topic that is badly presented in this course.
How can I describe this? The first two weeks are a chaotic overview of plotting in R. Three plotting systems are described but none of them is covered in enough detail. You could just skip the videos and Google how to plot different things. The last two weeks seem to be just filler material, clustering and dimensionality reduction appear out of the blue but there are no quizzes or assignments about those topics so you could as well skip them too. So at the end you have to do a couple of programming assignments about plotting in R with the help of Google and that's it. The examples and assignments are really non-interesting and unchallenging unless you are really into air pollution. It's bad, very bad.
This is a very nice course on solving difficult problems. The lectures are excellent very well presented and full of enthusiasm and humour. The instructors post frequently and monitor how the course is going. The programming assignments are very interesting, challenging and presented in a competitive way. There are only two things why I'm not giving this course 5 stars. The first thing is that the grading of the programming assigments is very coarse grained, you either get 0,3,7 or 10. It would be good to have the full range from 0 to 10 because not all 7's are the same. The second problem is that if you want to get full marks on the assignments you probably should use an external solver or library rather than code your own solution. I believe there should be a separate leaderboard for those that are coding your own solutions to learn the methods, otherwise you end up competing against the top libraries and programs in the industry. That's not a bad thing but it shouldn't affect your final grade.
This is a horrible course. The material is boring, presented in a confusing way and without any hint of enthusiasm or motivation. The course project is a total letdown, uninteresting and badly worded leading to a total chaos in peer reviewing. The course to me has been a total failure and the way it was presented very unprofessional.
The material is very bad, just a readout of slides presenting different aspects of R. The worst thing is the total lack of enthusiasm from the instructors and the boring examples used in the videos and the quizzes and programming assignments. Programming assignment #2 is particularly strange, a copy-paste challenge graded by peer reviewing. Not good at all.
The whole course can be completed in about 15 minutes. It's quite puzzling why they decided to start a specialization on such a bad way. Not really a course but just a way to ask you to install a couple of programs.