A Multi-View Learning Approach for Detecting Personality Disorders Among Arab Social Media Users

Abstract

Multi-view fusion approaches have gained increasing interest in the last few years by researchers. This interest comes due to the many perspectives that datasets can be looked at and evaluated. One of the most urging areas that require constant leveraging with latest technologies and multi-perspective judgments, is the area of Psychology. In this paper, a novel multi-view fusion model using deep learning algorithms is presented to detect popular types of Personality Disorders among the Arab users of Twitter Platform in an expert driven fashion, based on the descriptions of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). The work presented is the first of its kind with no publicly available datasets that report statements around Personality Disorders in the Arabic Language, thus we created AraPerson, a dataset which consists of 8000 textual tweets coupled with 8000 images that prescribe mental statuses for a total of 150 users collected with regular expressions generated under the supervision of domain experts. The dataset was fed into a baseline multi-view model by combining a CNN model with a Bi-LSTM model to detect two types of popular personality disorders by analyzing textual and visual posts on 150 users’ profiles. The experiments were followed with fusing DenseNet model with Bi-LSTM model, with experimenting the effect of using Concatenation, Addition, and Multiplication methods for vectors’ combination. The work presented in this paper is unprecedented, specifically in a controversial area such as personality disorders detection among Arab communities. The best reported accuracy is 87% which is very promising as the two types of personality disorders, which were investigated, are highly overlapping.

Machine Learning for Multimodal Mental Health Detection: A Systematic Review of Passive Sensing Approaches

“…com/affdex-for-market-research (accessed on 10 December 2023), DBFace https: //github.com/dlunion/DBFace (accessed on 10 December 2023), FaceMesh https://developers. google.com/android/reference/ com/google/mlkit/vision/ facemesh/FaceMesh (accessed on 10 December 2023), dlib [347] [ 25,53,55,56,68,79,80,83,91,92,94,98,99,101,108,174,193,199,201,203,220 OpenFace [269], Face++ [352],…”

“…BERT [266], ALBERT [297], XLNet [285], bidirectional gated recurrent unit (BiGRU) [341], itwiki (Italian Wikipedia2Vec model), Spanish model [399], EmoBERTa [298] (incorporate linguistic and emotional information), MiniLM [400] (supports multiple languages), GPT [401], TextCNN [402] [48,51,56,77,87,98,126,173,193,198,201] Weighted average or sum of modality-level scores N/A [51,68,147,198,200] Average confidence scores from lower-level prediction N/A [121] Combine predictions of individual modalities as inputs to secondary ML models SVM, decision tree, random forest, novel ML models [48,52,56,64,71,72,74,103,122,155,193] Hierarchical score/decision-level fusion Weighted voting fusion network [428] [122,195] Summation of question-level scores from rules enforced on modality-specific predictions N/A [88] Model level…”

“…Fully connected (FC) layers, multilayer perceptron (MLP), CNN, LSTM, BiLSTM, GRU, temporal convolutional network (TCN) [441] (with dilation for long sequences)-with activation function like Sigmoid, Softmax, ReLU, LeakyReLU, and GeLU Predict scores of assessment scales (regression) or probability distribution over classes (classification) [60,78,80,[84][85][86][87][88][90][91][92][93][94]96,98,105,111,113,117,131,133,135,136,[142][143][144]146,162,163,167,168,170,172,174,178,179,190,197,199,201,218,219,221,223,308] DCNN-DNN (combination of deep CNN and DNN), GCNN-LSTM (combination of gated convolutional neural network, which replaces a convolution block in CNN wi...…”