Title
Dynamic emotion modelling and anomaly detection in conversation based on emotional transition tensor.
Abstract
•A (CNN-LSTM) model is proposed for identifying the emotion of the conversation texts.•A novel method is proposed to model human’s personalities through conversations.•A dynamic emotion sampling method is proposed for assessing emotion transition.•A new method of abnormal emotion detection in human’s conversation is proposed.
Year
DOI
Venue
2019
10.1016/j.inffus.2018.04.001
Information Fusion
Keywords
Field
DocType
Hybrid deep learning model,Emotional transition,Anomaly detection,Social conversation
Anomaly detection,Conversation,Social media,Pattern recognition,Markov chain Monte Carlo,Tensor,Convolutional neural network,Sentiment analysis,Sampling (statistics),Artificial intelligence,Mathematics
Journal
Volume
ISSN
Citations 
46
1566-2535
3
PageRank 
References 
Authors
0.42
26
3
Name
Order
Citations
PageRank
Xiao Sun16419.23
Chen Zhang211241.68
Lian Li318940.80