Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/66643
Title: Visualizing relation between tags in StackOverflow
Authors: Teong, Ke Ming
Keywords: DRNTU::Engineering
Issue Date: 2016
Abstract: Tags have been used increasingly with the high usage of internet search engines. Tags can help searchers to narrow down the search spaces and obtain desired results with efficiency. There are a lot of tags in StackOverflow for the searchers to input during search. The aim of this project is to extract the relation between tags in StackOverflow. The knowledge of the relation between tags can be used for the research and implementation of exploratory search engines. Firstly, the category of each tag in StackOverflow will be extracted. This process includes three parts namely, preprocess, category extraction and postprocess. The extraction is completed using part of speech tagger and regular expression parser. Next, the relation and parent tag is retrieved. This process can be done using results from tag category and keyword searching. Lastly, the retrieved results will be written to csv file and displayed on a force directed graph using Firefox browser. In addition to the graph visualization, the accuracy of extracted category and relations had been investigated and evaluated. The accuracy test conducted was based on manual recognition for the correctness of each result. Two accuracy tests were done on each of the extracted results. The accuracy of extracted category was 75% on average while the accuracy of extracted relations was 91.5% on average.
URI: http://hdl.handle.net/10356/66643
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP_Report_Keming.pdf
  Restricted Access
1.13 MBAdobe PDFView/Open

Page view(s)

154
Updated on Jun 22, 2021

Download(s)

12
Updated on Jun 22, 2021

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.