Most Viewed Data Mining Talks at Videolectures

By Grant Marshall, Sept 2014

Today, we look at the top 25 most viewed data mining lectures on videolectures.net

The videos are taken from the most popular data mining videos on videolectures.net. These are the videos, including authors, length, and venue, sorted by views:

1. Big-Data Tutorial, Marko Grobelnik, 11250 views, 1:09:51 (July 4, 2012, at ESWC Summer School, Kalamaki 2012)
2. Statistical techniques for fraud detection, prevention, and evalutation, David J. Hand, 7836 views, 1:43:27 (December 3, 2007, at NATO Advanced Study Institute on Mining Massive Data Sets for Security, Gazzada 2007)
3. Mining Massive RFID, Trajectory, and Traffic Data Sets, Jiawei Han, Jae-Gil Lee, Hector Gonzalez, Xiaolei Li, 4912 views, 2:23:41 (September 26, 2008, at 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), Las Vegas 2008)
4. Data mining and Machine learning algorithms, Jose L. Balcazar, 4635 views, 2:15:01 (February 18, 2011, at AI Bootcamp 2011 @ AITI-KACE, Accra, Ghana)
5. From Trees to Forests and Rule Sets - A Unified Overview of Ensemble Methods, John Elder, Giovanni Seni, 4451 views, 2:44:39 (August 14, 2007, at 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), San Jose 2007)
6. Open Source Intelligence, Clive Best, 4407 views, 1:29:46 (December 3, 2007, at NATO Advanced Study Institute on Mining Massive Data Sets for Security)
7. Data Mining for Anomaly Detection, Aleksandar Lazarevic, Jaideep Srivastava, Vipin Kumar, Arindam Banerjee, Varun Chandola, 4172 views, 2:23:20 (October 10, 2008, at European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), Antwerp 2008)
8. Analysis of Time Series, Riccardo Bellazzi, 4145 views, 1:32:51 (February 25, 2007, at Advanced Course on AI (ACAI), Ljubljana 2005)
9. A Data Miner's Story - Getting to Know the Grand Challenges, Usama Fayyad, 4058 views, 0:48:33 (August 14, 2007, at 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), San Jose 2007)
10. Large-scale Data Mining: MapReduce and Beyond, Spiros Papadimitriou, Jimeng Sun, Rong Yan, 3751 views, 2:46:59 (October 1, 2010, at 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), Washington 2010)
11. Other ML/DM software (R, Weka, Yale), Lluis Belanche, 3417 views, 0:56:57 (July 5, 2007 at, PASCAL Bootcamp in Machine Learning, Vilanova 2007)
12. Data Mining, Rao Kotagiri, 2720 views, 3:06:58 (April 1, 2009, at Summer Schools in Logic and Learning, Canberra 2009)
13. Introduction to Graphical Models for Data Mining, Arindam Banerjee, 2068 views, 2:35:45 (October 1, 2010, at 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), Washington 2010)
14. The quantification of advertising and lessons from building a business based on large scale data mining, Konrad Feldman, 2067 views, 0:59:35 (October 1, 2010, at 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), Washington 2010)
15. Omrežja, Stanford in Silicijeva dolina, Jure Leskovec, 1600 views, 1:27:58 (January 7, 2013, recorded for VideoLectures.NET - Single Lectures Series)
16. Transfer learning in social recommendations, Qiang Yang, 1468 views, 0:42:14 (August 4, 2011, at International Workshop on Social Web Mining, Barcelona 2011)
17. Twitter Sentiment in Financial Domain, Miha Grcar, 1257 views, 0:16:55 (April 11, 2012, at FIRST Industrial Workshop, Milano 2012)
18. Scale-out Beyond MapReduce, Raghu Ramakrishnan, 1203 views, 0:51:49 (September 27, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
19. ORANGE: Data Mining Fruitful and Fun, Janez Demsar, 1195 views, 0:15:14 (November 16, 2012, at Conference on 100 Years of Alan Turing and 20 Years of SLAIS)
20. A Data Scientist's Guide to Making Money from Start-ups, Geoff Webb, Foster Provost, Ron Bekkerman, Oren Etzioni, Usama Fayyad, Claudia Perlich, 1169 views, 1:07:58 (September 27, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
21. Optimization in Learning and Data Analysis, Stephen J. Wright, 861 views, 1:01:24 (September 27, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
22. The Role of Information Diffusion in the Evolution of Social Network, Filippo Menczer, 547 views, 0:19:06 (September 27, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
23. Mining Heterogeneous Information Networks, Jiawei Han, 490 views, 0:30:30 (September 27, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
24. The Online Revolution: Education for Everyone, Andrew Ng, 381 views, 1:13:21 (September 20, 2013, at 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Chicago 2013)
25. Can cascades be predicted?, Jure Leskovec, 251 views, 1:03:49 (June 2, 2014, at Solomon seminar)
    
    

Looking at the content of the titles of these most popular videos, we may find hints about what the most popular topics in data mining among the viewers of videolectures.net. Based on this visualization, it seems lectures on MapReduce and Web Mining strike a chord with data mining lecture viewers.

One interesting facet of the data to look at is the length of the video compared to the number of views on the video. Below is a scatter plot mapping the rank of the video against its length.



There is a slight inverse correlation between the length of the video and its rank in views. The three major outliers to this general trend are Large-scale Data Mining: MapReduce and Beyond (10), Data Mining (12), and Introduction to Graphical Models for Data Mining (13). This means that, when it comes to data mining lectures, deeper content-filled lectures can be very popular, even if they take more time to develop.



Note that this only applies if we sort by the number of views. If we take the videos in the order sorted by videolectures.net’s popularity metric, there is no clear correlation between length and popularity. This is because the criteria for a “popular” video as defined by videolectures.net account for more factors than simply the number of views, potentially including features like votes and publishing date.

Related:

• Sep 2014 – Mar 2015 Meetings in Analytics, Big Data, Data Mining, and Data Science
• KDD 2013 videolectures: watch the top researchers in Data Mining and Knowledge Discovery
• Panel Report: A Data Scientist Guide to Startups