Probabilistic Semantic Analysis Research Articles

Corpus Microtext Analysis and English Teaching Model Innovation in Cloud Computing Background

Abstract The rapid development of cloud computing technology makes the English teaching corpus an excellent choice for innovative English teaching mode. This paper constructs a corpus of English teaching resources and English O2O teaching mode based on MOOC education platform, combines the probabilistic semantic analysis (PLSA) model with the potential semantic indexing to carry out microtext analysis of the English teaching corpus, and solves the PLSA model by using the expectation (EM) algorithm. For the English teaching corpus and English O2O teaching model constructed in this paper, a validation analysis was conducted in three aspects: micro text analysis, text lexical ability and teaching satisfaction. The results show that the PLSA model is able to obtain complete sentence expressions in the microtexture of the corpus when the weight coefficient of the lexical items is set to 0.3. The text vocabulary ability of students in English O2O teaching mode shows a significant difference level at 1% level, and the correlation coefficient of in-class to pre-class reaches 0.889. The English teaching corpus relying on MOOC platform and O2O teaching mode can effectively enhance students’ deep understanding of English sentences, improve students’ satisfaction with English teaching, and enhance students’ interest in learning English.

COMPLEX METHOD OF AUTOMATIC FONOSEMANTIC ANALYSIS OF TEXT INFORMATION ON THE BASIS OF EVALUATION OF HIGH SEMANTIC UNITS IN THE CONDITIONS OF INFORMATIVE CONTINUATION

The article proposes a method of analysis of text messages for revealing their semantic orientation and suggestive influence on the subconscious of a person, in the conditions of information confrontation. The method is based on the evaluation of the phonetic component of the word, using phonosemantic analysis. And also structural, probabilistic assessment and evaluation according to place in the sentence, using probabilistic semantic analysis. The method of automatic phonosemantic analysis is described, the algorithm scheme and the concept of the method of automatic semantic analysis are proposed on the basis of the analysis of the structure of the word, with the scheme of the algorithm. A complex method of phonosemantic analysis is proposed based on the estimation of meaningful semantic units, which seeks to combine the two methods described above and demonstrates an example of an estimation of the random, predicted semantically negative sentence. To demonstrate the estimates, a comparison of the results of the proposed method with the results of the general phonosemantic evaluation of the sentence was made. Based on the results, conclusions are formed.

A hybrid language model based on a recurrent neural network and probabilistic topic modeling

A language model based on features extracted from a recurrent neural network language model and semantic embedding of the left context of the current word based on probabilistic semantic analysis (PLSA) is developed. To calculate such embedding, the context is considered as a document. The effect of vanishing gradients in a recurrent neural network is reduced by this method. The experiment has shown that adding topic-based features reduces perplexity by 10%.

Including item characteristics in the probabilistic latent semantic analysis model for collaborative filtering

We propose a new hybrid recommender system that combines some advantages of collaborative and content-based recommender systems. While it uses ratings data of all users, as do collaborative recomme...

Unsupervised Learning by Probabilistic Latent Semantic Analysis

This paper presents a novel statistical method for factor analysis of binary and count data which is closely related to a technique known as Latent Semantic Analysis. In contrast to the latter meth...