Finding Users‘ Harass Need from Blog Articles and Recommending Solutions

  • 蔡 宗翰

Student thesis: Master's Thesis

Abstract

Modern humans inevitably encounter difficulties and frustrations of life but rarely teach and address these issues in the process of living and learning Unless the initiative to seek help or find the relevant books to read most people may use negative attitude to interpret difficulties and frustrations encountered These attitudes often could not resolve the matter but to make the issue even more to expand The rapid development of the Internet led to the rise of many social platform networking blog is one of them Blogs provide Internet users to communicate with each other to achieve the effect of text information exchange We are here to focus on the part of the users who write harass in their blog articles We observed that the user when writing the article will be a major Subject to represent the article Here we call it harass event According to the harass event would give rise to emotions we are here to call it harass emotion Harass event and harass emotion can trigger harass need and finally we will provide multiple solutions based on harass need to the user to solve problems in life Taking into account the issue of credibility we use books as solutions We think in contrast to the pages books have higher confidence We collect blog articles form PIXNET We use these articles to do observation、training and testing We also use Chinese segmentation and tagging tool to segment blog articles And then we can extract features of blog articles Then we can use these features to provide suitable solutions In part of solutions we cannot get the complete contents of the book Therefore we collect book information from online bookstore including book introduction user comment etc we use book information as a basis for the recommended solutions Experimental results show that our method can effectively identify user's need harass need Our method also can recommend books as effective solutions to help users solve problems in life
Date of Award2015 Aug 17
Original languageEnglish
SupervisorWen-Hsiang Lu (Supervisor)

Cite this

'