Knowledge Graph-Enabled Text-Based Automatic Personality Prediction

Abstract

How people think, feel, and behave primarily is a representation of their personality characteristics. By being conscious of the personality characteristics of individuals whom we are dealing with or deciding to deal with, one can competently ameliorate the relationship, regardless of its type. With the rise of Internet-based communication infrastructures (social networks, forums, etc.), a considerable amount of human communications takes place there. The most prominent tool in such communications is the language in written and spoken form that adroitly encodes all those essential personality characteristics of individuals. Text-based Automatic Personality Prediction (APP) is the automated forecasting of the personality of individuals based on the generated/exchanged text contents. This paper presents a novel knowledge graph-enabled approach to text-based APP that relies on the Big Five personality traits. To this end, given a text, a knowledge graph, which is a set of interlinked descriptions of concepts, was built by matching the input text’s concepts with DBpedia knowledge base entries. Then, due to achieving a more powerful representation, the graph was enriched with the DBpedia ontology, NRC Emotion Intensity Lexicon, and MRC psycholinguistic database information. Afterwards, the knowledge graph, which is now a knowledgeable alternative for the input text, was embedded to yield an embedding matrix. Finally, to perform personality predictions, the resulting embedding matrix was fed to four suggested deep learning models independently, which are based on convolutional neural network (CNN), simple recurrent neural network (RNN), long short-term memory (LSTM), and bidirectional long short-term memory (BiLSTM). The results indicated considerable improvements in prediction accuracies in all of the suggested classifiers.

Text-based automatic personality prediction using KGrAt-Net: a knowledge graph attention network classifier

“…Eventually, we are going to answer the research questions (as expressed in “ Introduction ” section) according to our findings as follows: RQ.1 The results of our investigations revealed that the knowledge graph attention network has a significant positive effect on text-based APP. Comparing the average accuracies between the first classification strategy proposed by KGrAt-Net and baseline models in Table 5 clearly denotes the fact that there are considerable differences between the average accuracies in almost all baseline models (except Ramezani et al 59 in which it has outperformed to KGrAt-Net in C, E, and A). Comparing the accuracies in each of the OCEAN traits which are depicted in Fig.…”

“… El-Demerdash et al 53 they have proposed a deep learning-based APP system that was based on data-level and classifier-level fusion, which exploits various levels of information to improve the performance of an APP system. Ramezani et al 59 they have proposed a knowledge graph-enabled APP system that classifies the input text by building, enriching, and embedding its corresponding knowledge graph. The Table 5 , provides an insight into the performance of baseline models in APP as well as the performance of proposed classification strategies by KGrAt-Net.…”

“…Ramezani et al 59 they have proposed a knowledge graph-enabled APP system that classifies the input text by building, enriching, and embedding its corresponding knowledge graph.…”

“…6 ), enriching the knowledge graph attention network by knowledge graph embedding, substantially enhances the text-based APP accuracies inasmuch as it concludes the best results among all baseline models. In fact, before the enrichment (as the results of KGrAt-Net’s first strategy illustrate), KGrAt-Net despite defeating all baseline models, failed to overcome one of the baseline models (Ramezani et al 59 ) in some traits (C, E, and A), which it is basically established on the basis of knowledge graph embedding. This obviously indicates that embedding the knowledge graph leads to benefit some neglected aspects of knowings that were missed by the attention mechanism.…”

“…The authors in Ref. 59 proposed a knowledge graph-enabled text-based APP system, relying on the Big Five model. Specifically, they first tried to build the corresponding knowledge graph for a given text using DBpedia knowledge base.…”