A Computational Linguistic Approach towards Understanding Wikipedia's Article for Deletion (AfD) Discussions
Abstract
With the thriving of online deliberation, Wikipedia's Article for Deletion (AfD) discussion has drawn a number of researchers' attention in the past decade. In this thesis we aim to solve two main problems: 1) how to help new users effectively participate in the discussion; and 2) how to make it efficient for administrators to make decision based on the discussion. To solve the first problem, we obtain a knowledge repository for new users by recognizing imperatives. We propose a method to detect imperatives based on syntactic analysis of the texts. And the result shows a good precision and reasonable recall. To solve the second problem, we propose a decision making support system that provides administrators with an reorganized overview of a discussion. We first divide the arguments in the discussion into several groups based on similarity; then further divide each group into subgroups based on sentiment (positive, neutral and negative). In order to classify sentiment polarity, we propose a recursive algorithm based on the dependency structure of the text. Comparing with the state of the art sentiment analysis tool by Stanford, our algorithm shows a promising result of 3-categories classification without requiring a large training dataset.