Please use this identifier to cite or link to this item:
https://hdl.handle.net/10356/73961
Title: | Mining in social media data : happiness forecast @ SG | Authors: | Tan, Poh Lian | Keywords: | DRNTU::Engineering::Computer science and engineering | Issue Date: | 2018 | Abstract: | Individual happiness in each region play a significant role for social metric. Happiness has often indirectly characterized and overshadowed by social media indicators. This project studies a methodology to measure the correlation between the real time expressions of individuals made across Singapore and range of social phenomena factors- population, dengue cluster and electricity consumption. We will examine the expression made on the social media -Twitter and uncover the happiness index over different regions. A total of 10,000 raw data in Twitter was collected which consists of users share thoughts, images, links for all the regions in Singapore. The collection of real-time tweets is customised to suit our project by using streaming API in Python. The next stage is to perform text-mining techniques to obtain the meaningful term. After data cleaning and pre-processing phrase, the parsed term will be tagged to a happiness index dictionary to compute the happiness scores (H-Score). Additionally, happiness index of singlish tokens will be further classified with Sentic API. Finally, we will be evaluating the relationships between the happiness scores and the real-world phenomena. | URI: | http://hdl.handle.net/10356/73961 | Schools: | School of Computer Science and Engineering | 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 | Size | Format | |
---|---|---|---|---|
AMENDED_FINAL_REPORT_TAN_POH_LIAN.pdf Restricted Access | 1.68 MB | Adobe PDF | View/Open |
Page view(s)
333
Updated on Mar 28, 2024
Download(s) 50
38
Updated on Mar 28, 2024
Google ScholarTM
Check
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.