This week I learned more about applying weights to documents through the use of an inverse document frequency in relation to term frequency. We discussed the equations involved with performing such a task along with parametric and zone indexes. For our programming assignment we needed to develop methods to exclude terms from being indexed as well as apply tf-idf weights to the terms.
• Describe your reactions to what you did.
I was very interested in the application of tf-idf, vector space modeling and cosine similarity. I didn’t initially understand what was going on with vector space modeling from the text but after viewing some videos on the subject I felt like I grasped the idea behind it and as a result better appreciated the advantage of applying cosine similarity. I have a hard time trying to fathom applying such a concept in code though. With that in mind I had a big let down this week as I don’t think I’ll be able to complete all of the tasks for the programming assignment. There are several aspects that I realize I have experience with from previous courses but I’ve had difficulty implementing. As a result, I’m anticipating (though I’m planning to work as long as I can on it) turning in my first incomplete assignment at UoPeople, which is very disheartening.
• …show more content…
Discuss how they were helpful.
My last programming assignment and learning journal feedback was pretty standard. Similarly, the responses to my discussion post were praiseful but not particularly constructive. You did pose a question regarding the efficiency of zones versus parametric index that made me think somewhat about why to use one, both or