A CLSTM-TMN for marketing intention detection. (May 2020)

• Record Type: Journal Article
• Authors: Wang, Yufeng; Ma, Kun; Garcia-Hernandez, Laura; Chen, Jing; Hou, Zhihao; Ji, Ke; Chen, Zhenxiang; Abraham, Ajith
• Is Part Of: Engineering applications of artificial intelligence. Volume 91(2020)
• Publication Date: 2020-05
• Subjects: Text classification -- Marketing intention -- Topic memory -- News; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2020.103595
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: In recent years, neural network-based models such as machine learning and deep learning have achieved excellent results in text classification. On the research of marketing intention detection, classification measures are adopted to identify news with marketing intent. However, most of current news appears in the form of dialogs. There are some challenges to find potential relevance between news sentences to determine the latent semantics. In order to address this issue, this paper has proposed a CLSTM-based topic memory network (called CLSTM-TMN for short) for marketing intention detection. A ReLU-Neuro Topic Model (RNTM) is proposed. A hidden layer is constructed to efficiently capture the subject document representation, Potential variables are applied to enhance the granularity of subject model learning. We have changed the structure of current Neural Topic Model (NTM) to add CLSTM classifier. This method is a new combination ensemble both long and short term memory (LSTM) and convolution neural network (CNN). The CLSTM structure has the ability to find relationships from a sequence of text input, and the ability to extract local and dense features through convolution operations. The effectiveness of the method for marketing intention detection is illustrated in the experiments. Our detection model has a more significant improvement in F1 (7%) than other compared models. Graphical abstract: Highlights: CLSTM-TMN. The CLSTM-TMN model for text classification based on TMN was proposed. CLSTM has the feature of storing historical information. The maximum pooling layer is built in the lower part of CNN to obtain the maximum value of feature map. When LSTM is used to obtain context information semantics, the problem of gradient disappearance can be solved. RNTM. A novel Neural Topic Model based on NTM is proposed. The modified linear unit layer in the model extracts topics, and the topics and documents are represented as a whole. This method is not easy to cause the loss of key information. The output of some neurons in RNTM is 0, which reduces the interdependence of parameters and the occurrence of overfitting.