Data Mining Research: Overview of 2007

The end of the year is soon here. It is thus time to make an overview of topics and discussions that have lead this blog during 2007. First, I want to thank each reader and each person commenting and therefore contributing the the life of this blog. I am also glad that many people subscribed to the feed through RSS or email notification. This is what I do with blogs I read. I read posts through my RSS feed reader and go to the blog only when I want to comment on a topic. Since Data Mining Research is not updated on a daily basis, I think RSS feed is a good alternative to direct visits.

Below, I summarized topics that lead DMR during 2007. The older (January 2007) come first:

• Plenty of new data mining blogs
• Limitations of data mining (finance, terrorism)
• Cluster validity (related to my thesis)
• Data mining book reviews
• SVM, neural networks and decision tree trends
• Feature selection (related to my thesis)
• DIKW hierarchy
• New Data Mining Search Engine

Moreover, certain posts were more popular than others. Here is a list of the five most popular posts of DMR for 2007 (according to the number of pageviews):

1. Data mining trends
2. Cluster validity: existing indices
3. SVM, neural network and decision tree
4. List of data mining blogs
5. Data mining useless in finance?

Feel free to give any feedback regarding the blog (what you like and what you hate). You can either comment this post or email me at sandro.saitta@gmail.com.

Comments