English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The discovery and recommendation of Chinese news hot subtopic is a growing research area, and currently technology in this area is not yet mature. This research, inspired by traditional LDA model, uses the key phrase as feature to construct a “bag-of-phrase”. On this basis, we propose the Chinese news hot subtopic discovery and recommendation method based on key phrase and the LDA model. Hot subtopic discovery and recommendation is the mainly innovation in this paper. During the cluster of hot subtopics, we select using the Longest Common Sequence (LCS) value as the similarity distance. To evaluate the proposed method, we used a mixed Chinese news dataset to experiment, and adopting: (1) time consumption of the training model, and (2) Perplexity value, and (3) quality of the discovery hot subtopic, three index to evaluate the performance of the method. The experimental results show that the proposed method can accurately discover the news hot subtopics, and also efficiently recommend relational hot subtopics in various fields. Introduction With the rapidly development of Internet and media technology, the influence of web news is increasing. Compared with traditional media news, web news score higher in terms of timeliness and convenience, and has gradually become the main medium for people to obtain useful information. In order to help users get valuable information, public opinion monitoring, information security, and other fields it has become a hot area of research. In traditional text mining, the vector space model (VSM) [1] is often used for document representation. On this basis, the researchers put forward the Latent Semantic Indexing (LSI) [2] method based on the co-occurrence of lexical entry rules to find the semantic relations between words. LSI comes with an immediate benefit of dimension reduction, but the low-frequency part of useful lexical entry lead to impaired performance. Topic modeling has lots of applications in the field of text mining and information retrieval [3]. The Latent Dirichlet Allocation (LDA) [4] model is an unsupervised probability generation model. The probability distribution on multiple potential topics is used to represent the document features, and each topic is represented by the probability distribution of the features. The method of parameter reasoning in LDA model usually uses Expectation propagation [5], VEM [6] and Gibbs Sampling among others. Radim [7] used LDA models to discover topics and implement engineering with python gensim tools package. Furthermore, a Locally Consistent Latent Dirichlet Allocation (LC-LDA) model was proposed to learn collective motion patterns through the use of low-level tracklet and bag-of-words features [8]. Yang et al. Proposed a novel approach based on a hierarchical latent topic model to learn and recognize scene and places [9]. Cha et al. proposed a method incorporating popularity into topic models for social network analysis [10]. Topic analysis can be further classified into two categories [11,12]. Hawes et al. [13] and Abbott et al. [14] have employed topic detection and tracking to extract signs of approval and disapproval from a corpus. Adams [15] reported a vector space model that produces documents and queries as eigenvectors and searches for topic corresponding to the content of conversation by computing similarity between vectors. An open-domain ICSI Meeting Corpus in 2008 emphasized the labeling and analysis of

Chinese News Hot Subtopic Discovery and Recommendation Method Based on Key Phrase and the LDA Model