Psycholinguistic Tripartite Graph Network for Personality Detection

Abstract

Most of the recent work on personality detection from online posts adopts multifarious deep neural networks to represent the posts and builds predictive models in a data-driven manner, without the exploitation of psycholinguistic knowledge that may unveil the connections between one's language usage and his psychological traits. In this paper, we propose a psycholinguistic knowledge-based tripartite graph network, TrigNet, which consists of a tripartite graph network and a BERT-based graph initializer. The graph network injects structural psycholinguistic knowledge from LIWC, a computerized instrument for psycholinguistic analysis, by constructing a heterogeneous tripartite graph. The graph initializer is employed to provide initial embeddings for the graph nodes. To reduce the computational cost in graph learning, we further propose a novel flow graph attention network (GAT) that only transmits messages between neighboring parties in the tripartite graph. Benefiting from the tripartite graph, TrigNet can aggregate post information from a psychological perspective, which is a novel way of exploiting domain knowledge. Extensive experiments on two datasets show that TrigNet outperforms the existing state-of-art model by 3.47 and 2.10 points in average F1. Moreover, the flow GAT reduces the FLOPS and Memory measures by 38% and 32%, respectively, in comparison to the original GAT in our setting.

TranSentGAT: A Sentiment-Based Lexical Psycholinguistic Graph Attention Network for Personality Prediction

“…To solve the problem of encoding long documents with sequential or hierarchical models that may misguide the personality detection models, Yang et al [38] presented a deep graph convolutional network that utilize learn to connect approach that learns to build the graphs. Also, TrigNet architecture which consists of a tripartite graph network and a BERT-based graph initializer was proposed by Yang et al [39] aims to structural psycholinguistic knowledge from LIWC for personality detection. In [40], the authors designed a system to capture the dependency information hidden in the psycholinguistic features combined with syntactic information of documents to predict personality prediction.…”

Orders Are Unwanted: Dynamic Deep Graph Convolutional Network for Personality Detection

“…The graph can be constructed statically or dynamically in NLP tasks (Wu et al 2021). The static approach constructs the graph during preprocessing, which leverages linguistic knowledge and manually defined rules such as dependency parse trees (Wang et al 2020), knowledge graphs (Xie et al 2020;Yang et al 2021c), and co-occurrence and document-word relations (Yao, Mao, and Luo 2019). A graph constructed by the static approach is usually fixed and the structure cannot be optimized during graph representation learning, which could be sub-optimal.…”