Text Retrieval and Search Engines course available at Coursera is done. It is actually first course I have fully completed with all the homework, quizzes etc.

My general impression is that

• Topics were very well explained by prof Cheng. Some parts were actually quite complex (like Probabilistic Retrieval Models)
• Quizzes were comprehensive and I think it was almost impossible to pass if you did’t follow lectures carefully, take notes etc. The bar was high since we had only one attempt to pass each quiz
• Participants were able pass the course either by doing quizzes or programming assignments. I didn’t try the second due to technology (C++ and MeTa package) – I simply didn’t want to waste time on refreshing language I’m not using at work.

I believe that one of the most important benefits one can get from studying theoretical background of commonly used tools (search engines) is to get familiar with state-of-the-art solutions of problems one might face – sometimes in different context. For example sometime ago I was involved into implementation of data quality algorithms for big integration platform. In most cases, we were able to generate a significant amount of measures describing some specific aspects of the processed data. The usual question was how to merge them together to gain full-picture overview of the data. The course left me with good understanding of how such problems were solved in systems that heavily relies on proper ranking similarity between complex objects.

In two moths Coursera will launch another course by prof. Cheng – this time regarding text mining https://www.coursera.org/course/textanalytics