Improving Topic Coherence Using Parsimonious Language Model and Latent Semantic Indexing

Dewangan, Jitesh Kumar; Sharaff, Aakanksha; Pandey, Sudhakar

doi:10.1007/978-981-15-1420-3_89

Cited by 5 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the number of topics is high, the generated model tends to overfit, so you must rely on more than just perplexity to judge a model. We will combine this with changes in topic coherence to determine (Dewangan et al 2020).…”

Section: Methodsmentioning

confidence: 99%

Thematic Analysis and Sentiment Evolution of Online Public Opinion on Hangzhou Asian Games: The Case of Sina Weibo

Che,

Gao,

2024

Preprint

View full text Add to dashboard Cite

It is of great significance to study the emotional evolution characteristics of significant sports events, to understand and master the changing law of online public opinion topics, to increase the attention and influence of sports events, to promote the dissemination of sports culture, and to correctly guide the direction of online public opinion on sports events. By dividing the life cycle of "Hangzhou Asian Games" microblog topics, using the LDA topic model and SnowNLP sentiment analysis, we mined the topics of online public opinion, analyzed the characteristics of public sentiment, and portrayed the sentiment evolution of "Hangzhou Asian Games" microblog topics in three dimensions: time, space and gender. The results show that Sina Weibo users are mainly optimistic about the topic of "Hangzhou Asian Games." The economically developed provinces in the southern and eastern regions are more optimistic about the topic than the economically less developed provinces in the western and northeastern regions, with higher user participation. In each life cycle, the emotional focus and emotional tendency show positive emotion>neutral emotion>negative emotion, and there are differences in the public opinion attention and emotional state of different topics in different periods, regions, and genders. Compared with other public opinion events, the emotional change of sports events public opinion is distinctive and unique, and its emotional evolution process is closely related to the development of different themes at different phases of sports events.

show abstract

Section: Methodsmentioning

confidence: 99%

Thematic Analysis and Sentiment Evolution of Online Public Opinion on Hangzhou Asian Games: The Case of Sina Weibo

Che,

Gao,

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The analysis used techniques in the field of machine learning and mining text such as word cloud [34,35], mapping latent variables with network connections [36,37], a bag of words [38,39], topic modeling by Latent Semantic Index (LSI) [39,40], emotion recognition (profiler by Plutchik emotions) [41,42] and geomapping [43].…”

Section: Methodsmentioning

confidence: 99%

Literature Review on Health Emigration in Rare Diseases—A Machine Learning Perspective

Skweres-Kuchta

Czerska

Szaruga

2023

IJERPH

View full text Add to dashboard Cite

The article deals with one of the effects of health inequalities and gaps in access to treatments for rare diseases, namely health-driven emigration. The purpose of the paper is to systematize knowledge about the phenomenon of health emigration observed among families affected by rare diseases, for which reimbursed treatment is available, but only in selected countries. The topic proved to be niche; the issue of “health emigration in rare diseases” is an area for exploration. Therefore, the further analysis used text mining and machine learning methods based on a database selected based on keywords related to this issue. The results made it possible to systematize the guesses made by researchers in management and economic fields, to identify the most common keywords and thematic clusters around the perspective of the patient, drug manufacturer and treatment reimbursement decision-maker, and the perspective integrating all the others. Since the topic of health emigration was not directly addressed in the selected sources, the authors attempted to define the related concepts and discussed the importance of this phenomenon in managing the support system in rare diseases. Thus, they indicated directions for further research in this area.

show abstract

“…LSI uses a singular value decomposition (SVD) of a large term-document matrix to identify a linear subspace such that the relationship between the term and document are captured. LSI was then adopted by various studies, such as Dewangan et al [23]. However, LSI is limited as the technique does not assign a probability to the topic.…”

Section: Related Workmentioning

confidence: 99%

Comparison of Topic Modelling Approaches in the Banking Context

et al. 2023

View full text Add to dashboard Cite

Topic modelling is a prominent task for automatic topic extraction in many applications such as sentiment analysis and recommendation systems. The approach is vital for service industries to monitor their customer discussions. The use of traditional approaches such as Latent Dirichlet Allocation (LDA) for topic discovery has shown great performances, however, they are not consistent in their results as these approaches suffer from data sparseness and inability to model the word order in a document. Thus, this study presents the use of Kernel Principal Component Analysis (KernelPCA) and K-means Clustering in the BERTopic architecture. We have prepared a new dataset using tweets from customers of Nigerian banks and we use this to compare the topic modelling approaches. Our findings showed KernelPCA and K-means in the BERTopic architecture-produced coherent topics with a coherence score of 0.8463.

show abstract

Improving Topic Coherence Using Parsimonious Language Model and Latent Semantic Indexing

Cited by 5 publications

References 11 publications

Thematic Analysis and Sentiment Evolution of Online Public Opinion on Hangzhou Asian Games: The Case of Sina Weibo

Thematic Analysis and Sentiment Evolution of Online Public Opinion on Hangzhou Asian Games: The Case of Sina Weibo

Literature Review on Health Emigration in Rare Diseases—A Machine Learning Perspective

Comparison of Topic Modelling Approaches in the Banking Context

Contact Info

Product

Resources

About