A Novel Sentence Embedding Based Topic Detection Method for Microblogs

Wei, Cong; Jiang, Shan; Wang, Cuirong

doi:10.1109/access.2020.3036043

Cited by 5 publications

(4 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It a technique that extracts various microblogging features they are emoticons, URLs, question marks, punctuation marks, all caps, hashtags, elongated words,user names, etc. Wan et al [ 45 ]. Hashtags: the number of hashtags.…”

Section: Proposed Methodology For Sentiment Analysis From Spam Smsmentioning

confidence: 99%

SMS sentiment classification using an evolutionary optimization based fuzzy recurrent neural network

Srinivasarao

Sharaff

2023

Multimed Tools Appl

View full text Add to dashboard Cite

Sentiment analysis using the inbox message polarity is a challenging task in text mining, this analysis is used to differentiate spam and ham messages in mail. Polarity estimation is mandatory for spam and ham identification, whereas developing a perfect architecture for such classification is the hot demanding topic. To fulfill that, fuzzy based Recurrent Neural network-based Harris Hawk optimization (FRNN-HHO) is introduced, which performs post-classification over the classified messages (spam and ham). Previously the authors tried to classify the spam and ham messages from the collection of SMSs. But sometimes, the spam messages may incorrectly be classified within the ham classes. This misclassification may reduce the accuracy. The sentiment analysis process is performed over the classified messages to improve such classification accuracy. The spam and ham messages from the available data are classified using a Kernel Extreme Learning Machine (KELM) classifier. The sentiment analysis and classification based experimental evaluation is carried out using accuracy, recall, f-measure, precision, RMSE, and MAE. The performance of the proposed architecture is evaluated using threedifferent datasets: SMS, Email, and spam-assassin. The Area under the curve (AUC) of the proposed approach is found to be 0.9699 (SMS dataset), 0.958 (Email dataset), and 0.95 (spam assassin).

show abstract

Section: Proposed Methodology For Sentiment Analysis From Spam Smsmentioning

confidence: 99%

SMS sentiment classification using an evolutionary optimization based fuzzy recurrent neural network

Srinivasarao

Sharaff

2023

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…There are also some researches on TDT around the improvement of classification or clustering algorithms. Wan et al [ 16 ] proposed a clustering algorithm RADBSCAN with relational awareness based on DBSCAN. This algorithm can determine the number of topics according to the dataset itself, and the topic recognition effect is better.…”

Section: Related Workmentioning

confidence: 99%

A Topic Recognition Method of News Text Based on Word Embedding Enhancement

Du¹,

Li²,

Liu³

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Topic recognition technology has been commonly applied to identify different categories of news topics from the vast amount of web information, which has a wide application prospect in the field of online public opinion monitoring, news recommendation, and so on. However, it is very challenging to effectively utilize key feature information such as syntax and semantics in the text to improve topic recognition accuracy. Some researchers proposed to combine the topic model with the word embedding model, whose results had shown that this approach could enrich text representation and benefit natural language processing downstream tasks. However, for the topic recognition problem of news texts, there is currently no standard way of combining topic model and word embedding model. Besides, some existing similar approaches were more complex and did not consider the fusion between topic distribution of different granularity and word embedding information. Therefore, this paper proposes a novel text representation method based on word embedding enhancement and further forms a full-process topic recognition framework for news text. In contrast to traditional topic recognition methods, this framework is designed to use the probabilistic topic model LDA, the word embedding models Word2vec and Glove to fully extract and integrate the topic distribution, semantic knowledge, and syntactic relationship of the text, and then use popular classifiers to automatically recognize the topic categories of news based on the obtained text representation vectors. As a result, the proposed framework can take advantage of the relationship between document and topic and the context information, which improves the expressive ability and reduces the dimensionality. Based on the two benchmark datasets of 20NewsGroup and BBC News, the experimental results verify the effectiveness and superiority of the proposed method based on word embedding enhancement for the news topic recognition problem.

show abstract

“…In the appendix we provide a detailed description of how we implemented this additional filter, along with a sample of text ranked from least to most related to economic concepts (Table 12). A number of recent works suggest using a similar approach where a final computational step is taken to classify text data, see Bao et al (2020), Wu et al (2020), Wan et al (2020) and Khan et al (2020). A small scale evaluation of the classification strategy (100 pictures and associated text) results in an estimated precision of 0.78 and estimated recall of 0.80.…”

Section: Selection Of Imagesmentioning

confidence: 99%