In the lab meeting on 27 November, Bruce will talk about co-training followed by “Single-View Co-training”, which is chapter 2 of Minmin Chen's thesis.
In the lab meeting on Wednesday, November 20th, Anoop and Mark are going to talk about the LSHTC shared task challenge, specifically about task 1. Task 1 is about large-scale multi-category classification of Wikipedia pages.
The details of the Task can be found on the lshtc website.
In the lab meeting tomorrow, Anoop will continue his tutorial on Tree-Adjoining Grammars. The slides can be found here.
In the lab meeting on Wednesday, Raheleh Makki from Dalhousie University will give a talk. A short description about the talk is below
One of the important factors in the performance of sentiment analysis methods is having a comprehensive sentiment lexicon. However, since sentiment words have different polarities not only in different domains, but also in different contexts within the same domain, constructing such context-specific sentiment lexicons is not an easy task. The high costs of manually constructing such lexicons motivate researchers to create automatic methods for finding sentiment words and assigning their polarities. However, existing methods may encounter ambiguous cases with contradictory evidences which are hard to automatically resolve. To address this problem, we aim to engage the user in the process of polarity assignment and improve the quality of the generated lexicon via minimal user effort. A novel visualization is employed to present the results of the automatic algorithm, i.e., the extracted sentiment pairs along with their polarities. User interactions are provided to facilitate the supervision process. The results of our user study demonstrate (1) involving the user in the polarity assignment process improves the quality of the generated lexicon significantly, and (2) participants in the study preferred our visual interface and conveyed that it is easier to use compared to a text-based interface.
The lab meeting will be from 1130 hours at TASC1 9408.
In the lab meeting on 30th October, Richard Frank and Piper Jackson will give a talk about their NLP-related project in Criminology. A short description is attached below:
In today’s world security is of great concern, and we frequently hear of violent extremists committing violent acts (shootings, for example). We, in the School of Criminology, are using a custom-built web-crawler to study extremist webpages for the purpose of recruitment. Currently webpages are evaluated and categorized manually for content, and there are a very large number of pages to evaluate. This specific research aims to use NLP and Semantic Analysis to (semi-)automatically evaluate each body of text to determine the topic and sentiment within the text. Eventually the goal is to automatically distinguish these violent extremist webpages from webpages discussing the same events from a security point of view (for example, a government site), or from a neutral point of view (for example, a news source reporting the event).
The talk will be from 1130 hours at TASC1 9408.