A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
A Novelty Analysis about an Impact of Tweets and Twitter Bios on Topic Quality Discovery using the Topic Modeling
https://orcid.org/http://orcid.org/0000-0001-7322-8653
Tweets and users are two key sources that characterize Twitter. Users have bios to describe their background and personal interests, as with tweet messages, furthermore analyzing the content of these tweets and the user's bio is the inspiration for this research. The topics discovered from tweets and bios are fairly conceivable alone, and the research challenge is how to measure the topics' quality once coupled. In this research, an attempt has been made in the novelty analysis of tweets and user's account bios by implementing topic models, i.e., Latent Dirichlet Allocation and Correlated Topic Model with the number of topics 10, 20, and 30, wherein hashtag and user pooling schemes have been applied to review tweets and bios that are topically equivalent, to receive documents that are not only convenient but also topically coherent, where better topics have been revealed. These tweets and bios are gathered from a Twitter user in a certain timeline using the hashtag #App. A set of dendrograms for bios and tweets text has been created to analyze the topics that have been rendered by topic modeling in order to build dendrograms and compare them. The entanglement value was determined after a visual comparison of the dendrograms. Between dendrograms, the cophenetic correlation coefficient has also been estimated. The findings showed that both the user bio and the tweet text had an impact on topic quality discovery. On the basis of numerous measurements conducted in this study, it has also been discovered that the LDA topic model outperforms the CTM topic model.
A Novelty Analysis about an Impact of Tweets and Twitter Bios on Topic Quality Discovery using the Topic Modeling
https://orcid.org/http://orcid.org/0000-0001-7322-8653
Tweets and users are two key sources that characterize Twitter. Users have bios to describe their background and personal interests, as with tweet messages, furthermore analyzing the content of these tweets and the user's bio is the inspiration for this research. The topics discovered from tweets and bios are fairly conceivable alone, and the research challenge is how to measure the topics' quality once coupled. In this research, an attempt has been made in the novelty analysis of tweets and user's account bios by implementing topic models, i.e., Latent Dirichlet Allocation and Correlated Topic Model with the number of topics 10, 20, and 30, wherein hashtag and user pooling schemes have been applied to review tweets and bios that are topically equivalent, to receive documents that are not only convenient but also topically coherent, where better topics have been revealed. These tweets and bios are gathered from a Twitter user in a certain timeline using the hashtag #App. A set of dendrograms for bios and tweets text has been created to analyze the topics that have been rendered by topic modeling in order to build dendrograms and compare them. The entanglement value was determined after a visual comparison of the dendrograms. Between dendrograms, the cophenetic correlation coefficient has also been estimated. The findings showed that both the user bio and the tweet text had an impact on topic quality discovery. On the basis of numerous measurements conducted in this study, it has also been discovered that the LDA topic model outperforms the CTM topic model.
A Novelty Analysis about an Impact of Tweets and Twitter Bios on Topic Quality Discovery using the Topic Modeling
https://orcid.org/http://orcid.org/0000-0001-7322-8653
Tweets and users are two key sources that characterize Twitter. Users have bios to describe their background and personal interests, as with tweet messages, furthermore analyzing the content of these tweets and the user's bio is the inspiration for this research. The topics discovered from tweets and bios are fairly conceivable alone, and the research challenge is how to measure the topics' quality once coupled. In this research, an attempt has been made in the novelty analysis of tweets and user's account bios by implementing topic models, i.e., Latent Dirichlet Allocation and Correlated Topic Model with the number of topics 10, 20, and 30, wherein hashtag and user pooling schemes have been applied to review tweets and bios that are topically equivalent, to receive documents that are not only convenient but also topically coherent, where better topics have been revealed. These tweets and bios are gathered from a Twitter user in a certain timeline using the hashtag #App. A set of dendrograms for bios and tweets text has been created to analyze the topics that have been rendered by topic modeling in order to build dendrograms and compare them. The entanglement value was determined after a visual comparison of the dendrograms. Between dendrograms, the cophenetic correlation coefficient has also been estimated. The findings showed that both the user bio and the tweet text had an impact on topic quality discovery. On the basis of numerous measurements conducted in this study, it has also been discovered that the LDA topic model outperforms the CTM topic model.
A Novelty Analysis about an Impact of Tweets and Twitter Bios on Topic Quality Discovery using the Topic Modeling
J. Inst. Eng. India Ser. B
Muthusami, Rathinasamy (author) / Saritha, Kandhasamy (author)
Journal of The Institution of Engineers (India): Series B ; 103 ; 1431-1441
2022-10-01
11 pages
Article (Journal)
Electronic Resource
English
Searching for Relevant Tweets Based on Topic-Related User Activities
| British Library Online Contents | 2016
Characterising “Green Building” as a topic in Twitter
| Emerald Group Publishing | 2019
Twitter Social Media Conversion Topic Trending Analysis Using Latent Dirichlet Allocation Algorithm
| DOAJ | 2022