An adaptive Latent Semantic Analysis for text mining

Tu, Hong; Phan, Tuoi Thi; Nguyen, Khu Phi

doi:10.1109/icsse.2017.8030943

Cited by 8 publications

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Learning topics or patterns from large corpus has drawn increasing attentions in data mining and related areas as more and more electronic document archives are available on the Internet. Recent researches in machine learning and text mining have developed many classical techniques, e.g., Latent Semantic Analysis (LSA) [1], [2], Probabilistic Latent Semantic Indexing (pLSI) [3], Latent Dirichlet Analysis (LDA) [4], and Topic Word Embedding [5] for finding patterns of words in large document collections. Among all those techniques, hierarchical probabilistic models, also known as ''topic model'', have become a widely used approach for exploratory and predictive analysis of text [6]- [10].…”

Section: Introductionmentioning

confidence: 99%

Mining Dynamics of Research Topics Based on the Combined LDA and WordNet

Feng

Zeng

et al. 2019

IEEE Access

View full text Add to dashboard Cite

A large volume of research documents are available online for us to access and analysis. It is very important to detect and mine the dynamics of the research topics from these large corpora. In this paper, we propose an improved method by introducing WordNet to LDA. This approach is to find latent topics of large corpora, and then we propose many methods to analyze the dynamics of those topics. We apply the methodology to two large document collections: 1940 papers from NIPS 00-13 (1987-2000) and 2074 papers from NIPS 14-23 (2001-2010). Six experiments are conducted on the two corpora and the experimental results show that our method is better than LDA in finding research topics and is feasible in discovering the dynamics of research topics. INDEX TERMS Research topics mining, dynamics of research topics, latent Dirichlet allocation, WordNet, large corpora.

show abstract

Section: Introductionmentioning

confidence: 99%

Mining Dynamics of Research Topics Based on the Combined LDA and WordNet

Feng

Zeng

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, in the traditional space vector model, there is no semantic relation between the words of vectors and each word needs to be vectorized, and the cosine calculation is performed to consume much time of processing of the Central Processing Unit. Based on the semantic similarity calculation method which is divided into the corpus based methods, the methods based on semantic dictionary and the method based on statistical language model [3][4][5][6][7][8], these methods are more complicated and are easily limited by the size of the corpus. The results of similarity calculation are more easily influenced by the noise of the training data.…”

Section: Introductionmentioning

confidence: 99%