A diversity-aware recommendation system for tutoring

PurposeImproving the performance of recommender systems (RSs) has always been a major challenge in the area of e-commerce because the systems face issues such as cold start, sparsity, scalability and interest drift that affect their performance. Despite the efforts made to solve these problems, there is still no RS that can solve or reduce all the problems simultaneously. Therefore, the purpose of this study is to provide an effective and comprehensive RS to solve or reduce all of the above issues, which uses a combination of basic customer information as well as big data techniques.Design/methodology/approachThe most important steps in the proposed RS are: (1) collecting demographic and behavioral data of customers from an e-clothing store; (2) assessing customer personality traits; (3) creating a new user-item matrix based on customer/user interest; (4) calculating the similarity between customers with efficient k-nearest neighbor (EKNN) algorithm based on locality-sensitive hashing (LSH) approach and (5) defining a new similarity function based on a combination of personality traits, demographic characteristics and time-based purchasing behavior that are the key incentives for customers' purchases.FindingsThe proposed method was compared with different baselines (matrix factorization and ensemble). The results showed that the proposed method in terms of all evaluation measures led to a significant improvement in traditional collaborative filtering (CF) performance, and with a significant difference (more than 40%), performed better than all baselines. According to the results, we find that our proposed method, which uses a combination of personality information and demographics, as well as tracking the recent interests and needs of the customer with the LSH approach, helps to improve the effectiveness of the recommendations more than the baselines. This is due to the fact that this method, which uses the above information in conjunction with the LSH technique, is more effective and more accurate in solving problems of cold start, scalability, sparsity and interest drift.Research limitations/implicationsThe research data were limited to only one e-clothing store.Practical implicationsIn order to achieve an accurate and real-time RS in e-commerce, it is essential to use a combination of customer information with efficient techniques. In this regard, according to the results of the research, the use of personality traits and demographic characteristics lead to a more accurate knowledge of customers' interests and thus better identification of similar customers. Therefore, this information should be considered as a solution to reduce the problems of cold start and sparsity. Also, a better judgment can be made about customers' interests by considering their recent purchases; therefore, in order to solve the problems of interest drifts, different weights should be assigned to purchases and launch time of products/items at different times (the more recent, the more weight). Finally, the LSH technique is used to increase the RS scalability in e-commerce. In total, a combination of personality traits, demographics and customer purchasing behavior over time with the LSH technique should be used to achieve an ideal RS. Using the RS proposed in this research, it is possible to create a comfortable and enjoyable shopping experience for customers by providing real-time recommendations that match customers' preferences and can result in an increase in the profitability of e-shops.Originality/valueIn this study, by considering a combination of personality traits, demographic characteristics and time-based purchasing behavior of customers along with the LSH technique, we were able for the first time to simultaneously solve the basic problems of CF, namely cold start, scalability, sparsity and interest drift, which led to a decrease in significant errors of recommendations and an increase in the accuracy of CF. The average errors of the recommendations provided to users based on the proposed model is only about 13%, and the accuracy and compliance of these recommendations with the interests of customers is about 92%. In addition, a 40% difference between the accuracy of the proposed method and the traditional CF method has been observed. This level of accuracy in RSs is very significant and special, which is certainly welcomed by e-business owners. This is also a new scientific finding that is very useful for programmers, users and researchers. In general, the main contributions of this research are: 1) proposing an accurate RS using personality traits, demographic characteristics and time-based purchasing behavior; 2) proposing an effective and comprehensive RS for a “clothing” online store; 3) improving the RS performance by solving the cold start issue using personality traits and demographic characteristics; 4) improving the scalability issue in RS through efficient k-nearest neighbors; 5) Mitigating the sparsity issue by using personality traits and demographic characteristics and also by densifying the user-item matrix and 6) improving the RS accuracy by solving the interest drift issue through developing a time-based user-item matrix.

Recommender systems have emerged, as an intelligent information filtering tool, to help users effectively identify information items of interest from a set of overwhelming choices and provide personalized services. Most recommendation technologies typically rely on ratings or item attributes to generate recommendations. Studies show that personality influences people's decision making processes and interests. However, little research has ventured into the area of incorporating personality into recommender systems. The utilization of personality characteristics in recommender systems and the exploration of user perceptions of personality-engaged recommendation technologies are the two main concerns of this thesis. In all of our studies, the five factor model (FFM), one of the most widely employed personality models, was adopted to establish user personality profiles. Firstly, we have aimed at designing effective personality-engaged recommender systems, with the emphasis on how to integrate personality into the recommendation generation framework, and how to utilize personality information to address the problems that exist in current recommender systems. We implemented a personality-based music recommender system prototype based on the findings from prior psychological studies. This system builds user personality profiles by means of personality quizzes, and it accordingly predicts their musical preferences and makes music recommendations. This system prototype was employed in our later user study investigating user perceptions of this novel recommendation technology. Moreover, we investigated how to incorporate personality into collaborative filtering recommender systems with the purpose of alleviating new user and dataset sparsity problems. In contrast to the traditional rating-based collaborative filtering method, the proposed three variants which take personality characteristics into account significantly improve the prediction accuracy in both cold-start scenarios (i.e., new user & dataset sparsity). Furthermore, in order to generalize the personality-engaged recommendation technology to other item domains, we used matrix factorization methods to automatically discover the links between personality traits and items. Consequently, recommendations were generated based on the discovered relationships. The empirical results show that the proposed methods achieve superior performances compared to other tested methods. In particular, the proposed method can produce highly accurate recommendations, even though no rating is available. It has been demonstrated that the proposed method can be applied to design effective gift recommender systems where rating information is hard to obtain. Secondly, we have conducted two user studies with the aim of investigating user perceptions of personality-engaged recommender systems. One compares a personality-quiz based movie recommender system with a baseline rating-based recommender system, and identifies the factors which lead to user acceptance to the personality-based system. The results show that the perceived accuracy in the two systems is not significantly different. However, users expended significantly less effort, both perceived cognitive effort and actual completion time, to establish their initial preference profiles in the personality quiz-based system than in the rating-based system. Additionally, users expressed stronger intentions to reuse the personality quiz-based system and introduce it to their friends. The other user study investigates the influence of contexts on the user perceptions of the personality-engaged recommender systems. We have examined two contextual factors. One is users' usage goals, finding items for the active user himself/herself or a friend as gifts. The other is the level of user domain knowledge. Our in-depth user studies show that while users perceived that the recommended items for their friends were more accurate, they enjoyed more using the personality-based recommender to find items for themselves than for their friends. Additionally, it has been found that the domain knowledge has a significant impact on user perceptions of the system. The results show that novice users, who are not knowledgeable about music, appreciated the personality-based recommender more than musical experts did. In the end, a set of design guidelines is derived from all of the experimental results. They should be helpful for designing satisfying and effective personality-engaged recommender systems.

In daily life, people usually rely on recommendations, traditionally given by other people (family, friends, etc.) for their most varied decisions. In the world, this is not different, given that recommender systems are present everywhere in such a way that we no longer realize. The main goal of these systems is to assist users in the decision-making process, generating recommendations that are of their interest and based on their tastes. These recommendations range from products in e-commerce websites, like books to read or places to visit to what to eat or how long one should walk a day to have a healthy life, who to date or who one should follow on social networks. And this is an increasing area. On the one hand, we have more and more users on the internet whose life is somewhat digitized, given than what one does in the world is represented in a certain way in the digital world. On the other hand, we suffer from information overload, which can be mitigated by the use of recommendation systems. However, these systems also face some problems, such as the cold start problem and their need to be more and more human, personalised and precise in order to meet the yearning of users and companies. In this challenging scenario, personality-based recommender systems are being increasingly studied, since they are able to face these problems. Some recent projects have proposed the use of the human personality in recommenders, whether as a whole or individually by facet in order to meet those demands. Therefore, this thesis is devoted to this new area of personality-based recommendation, focusing on one of its most important traits, the curiosity. Additionally, in order to exploit the information already present on the internet, we will implicitly obtain information from social networks. Thus, this work aims to build a better experience for the end user through a new approach that offers an option for some of the gaps identified in personality-based recommendation systems. Among these gap improvements, the use of social networks to feed the recommender systems soften the cold start problem and, at the same time, it provides valuable data for the prediction of the human personality. Another found gap is that the curiosity was not used by any of the studied recommender systems; almost all of them have used the overall personality of an individual through the Big Five personality traits. However, psychological studies confirm that the curiosity is a relevant trait in the process of choosing an item, which is directly related to recommendation systems. In summary, we believe that a recommendation system that implicitly measures the curiosity and uses it in the process of recommending new items, especially in the tourism sector, could clearly improve the capacity of these systems in terms of accuracy, serendipity and novelty, allowing users to obtain positive levels of satisfaction with the recommendations. This thesis begins with an exhaustive study of the state of the art, where we highlight works about recommender systems, the human personality from the point of view of traditional and positive psychology and how these aspects are combined. Then, we develop an online application capable of implicitly extracting information from the user profile in a social network, thus generating predictions of one or more personality traits. Finally, we develop the CURUMIM system, able to generate online recommendations with different properties, combining the curiosity and some sociodemographic characteristics (such as level of education) extracted from Facebook. The system is tested and assessed within the tourism context by real users. The results demonstrate its ability to generate novel and serendipitous recommendations, while maintaining a good level of accuracy, independently of the degree of curiosity of the users.

Understanding user’s behavior and their interactions with artificial-intelligent-based systems is as important as analyzing the performance of the algorithms used in these systems. For instance, in the Recommender Systems domain, the accuracy of the recommendation algorithm was the ultimate goal for most systems designers. However, researchers and practitioners have realized that providing accurate recommendations is insufficient to enhance users’ acceptance. A recommender system needs to focus on other factors that enhance its interactions with the users. Recent researches suggest augmenting these systems with persuasive capabilities. Persuasive features lead to increasing users’ acceptance of the recommendations, which, in turn, enhances users’ experience with these systems. Nonetheless, the literature still lacks a comprehensive view of the actual effect of persuasive principles on recommender users. To fill this gap, this study diagnoses how users of different characteristics get influenced by various persuasive principles that a recommender system uses. The study considers four users’ aspects: age, gender, culture (continent), and personality traits. The paper also investigates the impact of the context (or application domain) on the influence of the persuasive principles. Two application domains (namely eCommerce and Movie recommendations) are considered. A within-subject user study was conducted. The analysis of (279) responses revealed that persuasive principles have the potential to enhance users’ experience with recommender systems. The study also shows that, among the considered factors, culture, personality traits, and the domain of recommendations have a higher impact on the influence of persuasive principles than other factors. Based on the analysis of the results, the study provides insights and guidelines for recommender systems designers. These guidelines can be used as a reference for designing recommender systems with users’ experience in mind. We suggest that considering the results presented in this paper could help to improve recommender-users interaction.

Human personality traits are key drivers behind our decision making, influencing our lives on a daily basis. Inference of personality traits, such as the Myers-Briggs personality type, as well as an understanding of dependencies between personality traits and user behavior on various social media platforms, is of crucial importance to modern research and industry applications such as recommender systems. The emergence of diverse and cross-purpose social media avenues makes it possible to perform user personality profiling automatically and efficiently based on data represented across multiple data modalities. However, research efforts on personality profiling from multi-source multi-modal social media data are relatively sparse; the impact of different social network data on profiling performance and of personality traits on applications such as recommender systems is yet to be evaluated. Furthermore, large-scale datasets are also lacking in the research community. To fill these gaps, in this work we develop a novel multi-view fusion framework PERS that infers Myers-Briggs personality type indicators. We evaluate the results not just across data modalities but also across different social networks, and also evaluate the impact of inferred personality traits on recommender systems. Our experimental results demonstrate that PERS is able to learn from multi-view data for personality profiling by efficiently leveraging highly varied data from diverse social multimedia sources. Furthermore, we demonstrate that inferred personality traits can be beneficial to other industry applications. Among other results, we show that people tend to reveal multiple facets of their personality in different social media avenues. We also release a social multimedia dataset in order to facilitate further research on this direction.

Recommendation systems aim to provide end users with suggestions about items, social elements, products or services that are likely to be of their interests. Most studies on recommender systems focus on finding ways to improve the recommendations, including personalizing the systems based on details such as demographics, location, time and emotion, among others. In this work, a hybrid recommender system, namely HyPeRM, is presented, which uses users’ personality traits along with their demographic details (i.e. age and gender) to improve the overall quality of recommendations. The popular Big Five personality trait measurement scale was used to gauge users’ personalities. HyPeRM was evaluated using two metrics, that is, Standardized Root Mean Square Residual (SRMR) and Root Mean Square Error of Approximation (RMSEA). Both the metrics revealed that HyPeRM outperformed the baseline model (i.e. one without user’s personality) in terms of the recommendation accuracies. The study shows that user recommendations can be further enhanced when their personality traits are taken into consideration, and thus their overall search experience can be improved as well.

A recommendation system is an integral part of any modern online shopping or social network platform. The product recommendation system as a typical example of the legacy recommendation systems suffers from two major drawbacks: recommendation redundancy and unpredictability concerning new items (cold start). These limitations take place because the legacy recommendation systems rely only on the user's previous buying behavior to recommend new items. Incorporating the user's social features, such as personality traits and topical interest, might help alleviate the cold start and remove recommendation redundancy. Therefore, in this article, we propose Meta-Interest, a personality-aware product recommendation system based on user interest mining and metapath discovery. Meta-Interest predicts the user's interest and the items associated with these interests, even if the user's history does not contain these items or similar ones. This is done by analyzing the user's topical interests and, eventually, recommending the items associated with the user's interest. The proposed system is personality-aware from two aspects; it incorporates the user's personality traits to predict his/her topics of interest and to match the user's personality facets with the associated items. The proposed system was compared against recent recommendation methods, such as deep-learning-based recommendation system and session-based recommendation systems. Experimental results show that the proposed method can increase the precision and recall of the recommendation system, especially in cold-start settings.

In this study, we show that individual users’ preferences for the level of diversity, popularity, and serendipity in recommendation lists cannot be inferred from their ratings alone. We demonstrate that we can extract strong signals about individual preferences for recommendation diversity, popularity and serendipity by measuring their personality traits. We conducted an online experiment with over 1,800 users for six months on a live recommendation system. In this experiment, we asked users to evaluate a list of movie recommendations with different levels of diversity, popularity, and serendipity. Then, we assessed users’ personality traits using the Ten-item Personality Inventory (TIPI). We found that ratings-based recommender systems may often fail to deliver preferred levels of diversity, popularity, and serendipity for their users (e.g. users with high-serendipity preferences). We also found that users with different personalities have different preferences for these three recommendation properties. Our work suggests that we can improve user satisfaction when we integrate users’ personality traits into the process of generating recommendations.

With the development of recommendation systems, large amount of information collected from e-commerce could help customers to find the potential interesting products. Collaborative filtering and content-based recommendation systems are two common recommendation systems. While collaborative filtering has the problem of cold-start, content-based recommendation system could not explore the potential interests of users. Hybrid system combining these two techniques could achieve better results. This paper applies hybrid recommendation methods to the Amazon food reviews and evaluate the results in the aspects of precision, recall, diversity and novelty. It is found that the weighted hybrid recommendation system combing 0.95 weight of collaborative filtering and 0.05 content-based recommendation system achieves a good precision and diversity.

Genetic variation may play a significant role in the expression of complex personality and psychological traits. This article examines the relationship between heritable biological mechanisms and the psychological trait, impulsivity. In particular, dopamine is proposed to play a role in impulsive behaviours, and numerous studies have implicated functional polymorphisms of dopamine-related genes in impulsivity. This article reviews several studies concerning the role of dopamine receptor (DRD4) polymorphisms in the expression of an impulsivity sub-trait known as “novelty seeking”. Furthermore, this article focuses on recent approaches to the study of genetic variation, approaches to the measurement of novelty seeking, as well as other possible regulators of the trait in addition to genetics.

Cross-domain recommendation with user personality

Contextual information such as emotion, location and time can effectively improve product or service recommendations, however, studies incorporating them are lacking. This paper presents a context-aware recommender system, personalized based on several user characteristics and product features. The recommender system which was customized to recommend books, was aptly named as a Context-Aware Personalized Hybrid Book Recommender System, which utilized users’ personality traits, demographic details, location, review sentiments and purchase reasons to generate personalized recommendations. Users’ personality traits were determined using the Ten Item Personality Inventory. The results show an improved recommendation accuracy compared to the existing algorithms, and thus indicating that the integration of several filtering techniques along with specific contextual information greatly improves recommendations.

Personal trait is to measure the habitual patterns of behavior, thought, and emotion. It differs over individuals and is comparatively stable over time, relatively consistent over situations. Personal trait is significant for it has a lot of applications, such as recommendation system, chatbot and human resource management. It is convenient to recognize personal trait through wearable devices, social media and so on. Traditionally, personal trait is measured in general categories such as Big Five, which contains five traits: extroversion, neuroticism, agreeableness, conscientiousness, and openness. However, it is too abstract to describe personal trait in five aspects. We need the personal trait measured in more specific aspects, such as trait of interest or affect. We can know a person better through the traits in specific aspects than in the traditional abstract ways. In this paper, we proposed a general method of measuring personal trait called Personal Trait Matrix including topic word extraction and the word representation by word2vec based on user-generated text. Then a case study is made with datasets called myPersonality. The diversity of affects and social interactions are measured. Next, the correlation between the trait and the personality of Big Five was analyzed and discussed. The results demonstrate that the proposed method can measure the personal trait in affect and social interactions.

As governed by personality trait theory, humans tackle problems differently depending on their long-term behavioral characteristics. Computational awareness of personality traits fuels affective computing research, which investigates how to reliably recognize and utilize personality traits. Applications are diverse, including therapy monitoring, learning assistance, and recommender systems. Data-driven approaches are a promising path forward towards personality-aware human-computer interactions. Thereby, central challenges are the non-disruptive data acquisition, the time frame over which data must be collected before predictions become accurate, and the feature-centered data reduction to train reliable and lightweight machine learning models. In this work, we address these challenges by presenting a fully-automatic feature extraction and machine learning pipeline that makes accurate personality trait predictions for the widely-used Five Factor Model from passively-collected, short-term smartphone typing data collected from 76 participants (68 university students) in the wild. Our model allows for personality trait assessments after one day of data collection, demonstrating that, despite being a long-term behavioral trend, personality traits can be inferred accurately from shorter time periods. We demonstrate that our system can accurately predict personality traits on two levels (low and high) with up to 74.5% accuracy and 0.72 AUC for a single day, and up to 84.5% accuracy and 0.79 AUC after subsequent refinement over 10 weeks.

Mining personality traits from social messages for game recommender systems