Following the rectangular tables of relational databases and the triangular trees of semistructured data, the remaining Inf1-DA lectures will address the representation and analysis of more unstructured data. Today’s lecture provided a brief introduction to the classic information retrieval task of searching a large collection of documents to find those that match a simple query.
The focus here is not on specific algorithms or data representations, but on specifying the problem, how to recognise when you have a solution, and how to rate the performance of different competing solutions. In this case that means distinguishing between precision and recall in information retrieval; considering how each might be important in different problem domains; and the use of blends like the F-score to combine both measures.
The lecture finished with material on IBM’s Watson system using all kinds of data and analysis, including information retrieval, to perform question-answering on Jeopardy!.
University server disconnection today means that Panopto did not record the lecture; and there are also no recordings of this specific topic from previous years. My apologies. Instead I can offer the following:
-
The videos of IBM Watson used in the lecture.
-
Bix Beiderbecke performing Sorry which was, coincidentally, today’s opening music.
Links: Slides for Lecture 15
Homework
Watch These
What is Watson? The Science Behind an Answer
IBM Watson and Cancer Treatment Recommendations
Read These
IBM Watson Cognitive Cooking Fact Sheet
Watson in the Wild: March 5th 2015
References
These and more IBM Watson videos are available on the Inf1-DA 2017 IBM Watson YouTube playlist.
Cognitive Cookery
Say what foods you like and name some ingredients: Chef Watson will invent a recipe just for you: https://www.ibmchefwatson.com
Read what New Scientist reviewers thought about it (“occasionally inspired”; “one recipe called for precisely 554 juniper berries”). You can even buy the book.
