Personality Prediction of Social Network Users Using Ensemble and XGBoost
01 Jan 2020-pp 133-140
TL;DR: Predicting personality using social media is a new approach where direct interaction with people can be eliminated and accurate predictions can be built and high accuracy of 82.59% with an Ensemble is indicated.
Abstract: Machine learning has gained tremendous attention from researchers recently. It has wide applications in tasks such as prediction and classification. Current work focuses on the effective detection of the personality of social network users. Personality is a combination of one’s thinking and behavior. Having knowledge about personality of a person has many applications in real world such as varied recommendation systems or HR departments. Personality of a person can be better understood by interacting with him/her. Predicting personality using social media is a new approach where direct interaction with people can be eliminated and accurate predictions can be built. Although different machine learning methods have been used by researchers recently for the task of prediction, the use of Ensembles has not been explored. Current work focuses on advanced classifiers such as XGBoost and Ensemble for prediction. Experimentation on the real-time Twitter dataset indicates high accuracy of 82.59% with an Ensemble. These results are encouraging for future research.
Citations
More filters
01 Aug 2020
TL;DR: The aim of this research is to predict the personality of user by using the status information present in their social media profile to set up a framework that can predict the individual's personality based on Facebook user details.
Abstract: Social network usage is growing exponentially every day. Different information are typically exchanged via social media platforms such as Facebook. User knowledge and what they have conveyed through changes in status are useful for learning about the behavior and human personality assessment. This work aims at setting up a framework that can predict the individual's personality based on Facebook user details. In order to analyze the individual's personality, big five model is used. The aim of this research is to predict the personality of user by using the status information present in their social media profile. Based on the analysis result, the user's personality is further classified into one of the categories present in the OCEAN model. The accuracy of personality prediction achieved by using Random Forest Classifier is 64.25%. The mean squared error is achieved using random forest regressor is 5.25.
11 citations
Cites methods from "Personality Prediction of Social Ne..."
...In addition to approaches based on the development of different classifiers from different sub-sets of data or different sub-sets of features (bagging, random forests) or methods based on the combination of poor learners (boosting), it is also possible to create meta-learners (sometimes called stacked learning) who learn to make predictions based on input features[10]....
[...]
01 Jan 2021
TL;DR: It is indicative that a combination of fast recurrent neural networks and CNN may produce high accuracy with minimum time complexity, as existing researchers reflect CNN provides around 96.50% average accuracy for sentiment classification on the flicker image dataset.
Abstract: Determining the image sentiment is a tedious task for classification algorithms, owing to complexities in the raw images as well as the intangible nature of human sentiments. Classifying image sentiments is an evergreen research area, especially in social data analytics. In current times, it is a common practice for majority people to precise their feelings on the web by substituting text with the upload of images via a multiplicity of social media sites like Facebook, Instagram, Twitter as well as any other platform. To identify the emotions from visual cues, some visual features as well as image processing techniques are used. Several existing systems have already introduced emotion detection using machine learning techniques, but the traditional feature extraction strategies do not achieve the required accuracy on random objects. In the entire process, normalization of image, feature extraction, and feature selection are important tasks in the train module. This work articulates the newest developments in the field of image sentiment employing deep learning techniques. Also, the use of conventional machine learning techniques is compared along with deep learning algorithms. It is indicative that a combination of fast recurrent neural networks and CNN may produce high accuracy with minimum time complexity. It is noted from the survey that existing researchers reflect CNN provides around 96.50% average accuracy for sentiment classification on the flicker image dataset.
10 citations
TL;DR: A novel knowledge graph-enabled approach to text-based APP that relies on the Big Five personality traits is presented, which indicated considerable improvements in prediction accuracies in all of the suggested classifiers.
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.
7 citations
01 Jan 2021
TL;DR: This study endeavored to build a system that could predict an individual's personality through SM conversation using six supervised machine learning algorithms to handle unstructured and unbalanced SM conversations.
Abstract: The content of social media (SM) is expanding quickly with individuals sharing their feelings in a variety of ways, all of which depict their personalities to varying degrees. This study endeavored to build a system that could predict an individual's personality through SM conversation. Four BIG5 personality items (i.e. Extraversion (EXT), Consciousness (CON), Agreeable (AGR) and Openness to Experiences (OPN) equivalent to the Myers–Briggs Type Indicator (MBTI)) were predicted using six supervised machine learning (SML) algorithms. In order to handle unstructured and unbalanced SM conversations, three feature extraction methods (i.e. term frequency and inverse document frequency (TF-IDF), the bag of words (BOW) and the global vector for word representation (GloVe)) were used. The TF-IDF method of feature extraction produces 2–9% higher accuracy than word2vec representation. GloVe is advocated as a better feature extractor because it maintains the spatial information of words.
5 citations
TL;DR: This work surveys several supervised ML algorithms for identifying FLR frequencies by using measurements of the European quasi‐Meridional Magnetometer Array and evaluates the algorithm performance on four different station pairs, showing that tree‐based algorithms are robust and accurate models to achieve this goal.
4 citations
Cites methods from "Personality Prediction of Social Ne..."
...More in detail, XGB is a recent algorithm resulted in being highly efficient in regression/classification competition (see www.kaggle.com) and many real-world applications (Ivanov et al., 2020; Kunte & Panicker, 2020; Luckner et al., 2017)....
[...]
References
More filters
TL;DR: Results show that the predictive power of digital footprints over personality traits is in line with the standard “correlational upper-limit” for behavior to predict personality, with correlations ranging from 0.29 (Agreeableness) to 0.40 (Extraversion).
270 citations
TL;DR: A comparative analysis of state-of-the-art computational personality recognition methods on a varied set of social media ground truth data from Facebook, Twitter and YouTube is performed.
Abstract: A variety of approaches have been recently proposed to automatically infer users' personality from their user generated content in social media. Approaches differ in terms of the machine learning algorithms and the feature sets used, type of utilized footprint, and the social media environment used to collect the data. In this paper, we perform a comparative analysis of state-of-the-art computational personality recognition methods on a varied set of social media ground truth data from Facebook, Twitter and YouTube. We answer three questions: (1) Should personality prediction be treated as a multi-label prediction task (i.e., all personality traits of a given user are predicted at once), or should each trait be identified separately? (2) Which predictive features work well across different on-line environments? and (3) What is the decay in accuracy when porting models trained in one social media environment to another?
173 citations
TL;DR: This study attempts to build a system that can predict a person’s personality based on Facebook user information by implementing some deep learning architectures and succeeds to outperform the accuracy of previous similar research.
113 citations
TL;DR: A new approach that uses the social media platform Twitter to quantify suicide warning signs for individuals and to detect posts containing suicide-related content and the application of the martingale framework highlights changes in online behavior and shows promise for detecting behavioral changes in at-risk individuals.
Abstract: Suicidal ideation detection in online social networks is an emerging research area with major challenges. Recent research has shown that the publicly available information, spread across social media platforms, holds valuable indicators for effectively detecting individuals with suicidal intentions. The key challenge of suicide prevention is understanding and detecting the complex risk factors and warning signs that may precipitate the event. In this paper, we present a new approach that uses the social media platform Twitter to quantify suicide warning signs for individuals and to detect posts containing suicide-related content. The main originality of this approach is the automatic identification of sudden changes in a user's online behavior. To detect such changes, we combine natural language processing techniques to aggregate behavioral and textual features and pass these features through a martingale framework, which is widely used for change detection in data streams. Experiments show that our text-scoring approach effectively captures warning signs in text compared to traditional machine learning classifiers. Additionally, the application of the martingale framework highlights changes in online behavior and shows promise for detecting behavioral changes in at-risk individuals.
107 citations
TL;DR: The impact of the Big Five personality traits on human online information seeking is explored and individuals high in conscientiousness performed fastest in most information-seeking tasks, followed by those high in agreeableness and extraversion.
Abstract: We studied eye-movement behavior in different information-seeking tasks.We found three patterns of information seeking based on users personality facets.Personality traits drive information seeking differently depending on the task.Eye-movement parameters can predict these patterns in different information-seeking behaviors. Although personality traits may influence information-seeking behavior, little is known about this topic. This study explored the impact of the Big Five personality traits on human online information seeking. For this purpose, it examined changes in eye-movement behavior in a sample of 75 participants (36 male and 39 female; age: 2239 years; experience conducting online searches: 512 years) across three types of information-seeking tasks factual, exploratory, and interpretive. The International Personality Item Pool Representation of the NEO PI-R (IPIP-NEO) was used to assess the participants personality profile. Hierarchical cluster analysis was used to categorize participants based on their personality traits. A three cluster solution was found (cluster one consists of participants who scored high in conscientiousness; cluster two consists of participants who scored high in agreeableness; and cluster three consists of participants who scored high in extraversion). Results revealed that individuals high in conscientiousness performed fastest in most information-seeking tasks, followed by those high in agreeableness and extraversion. This study has important practical implications for intelligent human computer interfaces, personalization, and related applications.
94 citations