The Paper of the Week site is exactly as it sounds, a paper posted once a week for anyone interested to read. I will post some type of journal, conference, white paper, article (actually, I will just post a link to it, since it would most likely violate copyrights to post the actual paper) every week (starting Monday, January 8th) that I am going to read and then post comments on my understanding of the paper. I may not get it right, I may struggle to understand it, I may think it doesn’t make much sense, I may think it is right on. My goal is to fully understand each paper by the end of the week, file it away somewhere in the depths of my brain and move on. I would love to hear from others who get it (even the authors) and who want to explain it in laymen terms (which I will try to do.)

In the beginning, most papers will be in the Computer Science (CS) field, and will most likely have to do with Information Retrieval (IR), Machine Learning (ML), Natural Language Processing (NLP), Question Answering (QA), CS Algorithms and Data Structures. I encourage suggestions for articles related to these fields.

Who am I and why these areas of interest you ask? My name is Grant Ingersoll, and I currently work for a small startup. I am also a committer on the Lucene Java (see my Lucene blog at http://lucene.grantingersoll.com) project at Apache. As for the why part, it is part New Year’s resolution, part something I’ve wanted to do for a while. At CNLP I work with some really smart people who are very talented researchers. While I consider myself a pretty good engineer, who can code up most anything, I’ve always felt like I have some gaps in my background knowledge of the field (especially machine learning) that I wanted to fill in, so this is my public attempt to do so. I figure by posting it and commenting on it, I will cement my understanding of the problem discussed. If you all come along and join in, all the better, as we will all hopefully learn from it.

As for the paper selection process, it will most likely involve doing searches on Google Scholar for the fields described above and then looking into the “big” papers first, i.e. the ones with the most citations. I am open to suggestions, however, so please post comments on a paper you would like to see. The first paper will be announced next Monday, January 8th. Hope to see you then.

Popularity: 3% [?]