Serendipitous Recommendations Research Articles

Deep Learning Models for Serendipity Recommendations: A Survey and New Perspectives

Serendipitous recommendations have emerged as a compelling approach to deliver users with unexpected yet valuable information, contributing to heightened user satisfaction and engagement. This survey presents an investigation of the most recent research in serendipity recommenders, with a specific emphasis on deep learning recommendation models. We categorize these models into three types, distinguishing their integration of the serendipity objective across distinct stages: pre-processing, in-processing, and post-processing. Additionally, we provide a review and summary of the serendipity definition, available ground truth datasets, and evaluation experiments employed in the field. We propose three promising avenues for future exploration: (1) leveraging user reviews to identify and explore serendipity, (2) employing reinforcement learning to construct a model for discerning appropriate timing for serendipitous recommendations, and (3) utilizing cross-domain learning to enhance serendipitous recommendations. With this review, we aim to cultivate a deeper understanding of serendipity in recommender systems and inspire further advancements in this domain.

Interactive Feedback Loop with Counterfactual Data Modification for Serendipity in a Recommendation System

Users often encounter tedious recommendations as they are continuously exposed to the recommendation system. In response to this issue, serendipity in a recommendation system has been introduced to generate novel and unexpected recommendations while keeping them relevant to users’ previous preferences. This study proposes an interactive feedback loop for a serendipity in a recommendation system that allows users to directly explore content via counterfactual manipulation of features. Specifically, users indicate their preferences through the “what-if” based customization of content meta-information, and these modifications influence their usage history, thereby enabling the elicitation of serendipitous items. To validate the proposed feedback loop, we conducted a scenario-based experiment and compared system-initiated and user-intervened recommendations. The results reveal that counterfactual exploration can help to generate serendipitous recommendations. This study contributes to providing a user-friendly recommendation system that can retrieve preference-reflected recommendations through user interaction.

Deep neural network approach for a serendipity-oriented recommendation system

Most of the available recommender systems focus on the accuracy of recommendations. As a result, their recommendations are often popular and very close to user preferences, which make them repetitious and predictable, hence adversely affecting user satisfaction. Recent studies on recommender systems, however, aim for factors beyond accuracy as accuracy alone cannot ensure the satisfaction of all users. One of the most important criteria beyond the accuracy is serendipity, which includes relevant, unexpected, and novel recommendations that cannot be easily discovered by users themselves. In this paper, a Convolutional Neural Network (CNN) is integrated with the Particle Swarm Optimization (PSO) algorithm to generate serendipitous recommendations. The proposed method is based on the focus shift points, consisting of unexpectedness and relevance parameters. In this approach, these points are considered as the factors showing whether recommendations are serendipitous. The CNN is employed to predict the focus shift points for each user. Then, the PSO is utilized to search for recommendations close to the predicted focus shift points and generate the list of candidate recommendations. After that, the Serendipitous Personalized Ranking (SPR) method is employed to re-rank the candidate recommendations and generate the final list. According to the evaluation results, the proposed approach outperforms other state-of-the-art methods in SRDP, Hit Ratio, and NDCG factors.

Grace announces 2021 FCC catalysts pricing actions

Most of the available recommender systems focus on the accuracy of recommendations. As a result, their recommendations are often popular and very close to user preferences, which make them repetitious and predictable, hence adversely affecting user satisfaction. Recent studies on recommender systems, however, aim for factors beyond accuracy as accuracy alone cannot ensure the satisfaction of all users. One of the most important criteria beyond the accuracy is serendipity, which includes relevant, unexpected, and novel recommendations that cannot be easily discovered by users themselves. In this paper, a Convolutional Neural Network (CNN) is integrated with the Particle Swarm Optimization (PSO) algorithm to generate serendipitous recommendations. The proposed method is based on the focus shift points, consisting of unexpectedness and relevance parameters. In this approach, these points are considered as the factors showing whether recommendations are serendipitous. The CNN is employed to predict the focus shift points for each user. Then, the PSO is utilized to search for recommendations close to the predicted focus shift points and generate the list of candidate recommendations. After that, the Serendipitous Personalized Ranking (SPR) method is employed to re-rank the candidate recommendations and generate the final list. According to the evaluation results, the proposed approach outperforms other state-of-the-art methods in SRDP, Hit Ratio, and NDCG factors.

Serendipity in the city: User evaluations of urban recommender systems

AbstractThe contemporary city is increasingly being labeled as a smart city consisting of both physical and virtual spaces. This digital augmentation of urban life sets the scene for urban recommender systems to help citizens dealing with the abundance of digital information and corresponding choice overload, for example, by recommending the best place to have dinner based on your personal profile. There are, however, concerns that this kind of algorithmic filtering could lead to homogenization of urban experiences and a decline of social cohesion among citizens. To overcome this issue, scholars increasingly encourage the introduction of serendipity in all types of recommender systems. Nonetheless, it remains unclear how this can be achieved in practice. In this work, we study user evaluations of serendipity in urban recommender systems through a survey among 1,641 citizens. More specifically, we study which characteristics of recommended items contribute to serendipitous experiences and to what extent this increases user satisfaction and conversion. Our results align with findings in other application domains in the sense that there is a strong relation between the relevance and novelty of recommendations and the corresponding experienced serendipity. Moreover, serendipitous recommendations are found to increase the chance of users following up on these recommendations.

Serendipity in Recommender Systems: A Systematic Literature Review

A recommender system is employed to accurately recommend items, which are expected to attract the user’s attention. The over-emphasis on the accuracy of the recommendations can cause information over-specialization and make recommendations boring and even predictable. Novelty and diversity are two partly useful solutions to these problems. However, novel and diverse recommendations cannot merely ensure that users are attracted since such recommendations may not be relevant to the user’s interests. Hence, it is necessary to consider other criteria, such as unexpectedness and relevance. Serendipity is a criterion for making appealing and useful recommendations. The usefulness of serendipitous recommendations is the main superiority of this criterion over novelty and diversity. The bulk of studies of recommender systems have focused on serendipity in recent years. Thus, a systematic literature review is conducted in this paper on previous studies of serendipity-oriented recommender systems. Accordingly, this paper focuses on the contextual convergence of serendipity definitions, datasets, serendipitous recommendation methods, and their evaluation techniques. Finally, the trends and existing potentials of the serendipity-oriented recommender systems are discussed for future studies. The results of the systematic literature review present that the quality and the quantity of articles in the serendipity-oriented recommender systems are progressing.

Trust and fuzzy inference based cross domain serendipitous item recommendations (TFCDSRS)

Recommender System (RS) is an information filtering approach that helps the overburdened user with information in his decision making process and suggests items which might be interesting to him. While presenting recommendation to the user, accuracy of the presented list is always a concern for the researchers. However, in recent years, the focus has now shifted to include the unexpectedness and novel items in the list along with accuracy of the recommended items. To increase the user acceptance, it is important to provide potentially interesting items which are not so obvious and different from the items that the end user has rated. In this work, we have proposed a model that generates serendipitous item recommendation and also takes care of accuracy as well as the sparsity issues. Literature suggests that there are various components that help to achieve the objective of serendipitous recommendations. In this paper, fuzzy inference based approach is used for the serendipity computation because the definitions of the components overlap. Moreover, to improve the accuracy and sparsity issues in the recommendation process, cross domain and trust based approaches are incorporated. A prototype of the system is developed for the tourism domain and the performance is measured using mean absolute error (MAE), root mean square error (RMSE), unexpectedness, precision, recall and F-measure.

NASA-TLX–based workload assessment for academic resource recommender system

Recommender systems are expected to promote a student-centered teaching and learning environment. The age of information abundance has proven to need such systems. Recommender systems have been used to recommend learning items related to students’ research interests. Serendipity has also made its way into the academic environment, as systems recommend items that are useful and surprising to learners. Understanding user workload is important for students who use serendipitous recommender systems. In this research, we investigate various user interfaces for academic recommender systems by looking at students who are attempting to obtain serendipitous recommendations for their academic tasks. The study was evaluated on the NASA task load index (NASA-TLX). Our priority was to understand the mental, physical, and other workload attributes that can change when students seek serendipitous recommendations. We studied Mendeley, Google Scholar, Academia.edu, and ResearchGate. Our study found no substantial serendipitous recommendations observed by the users, but a few traces of serendipitous experiences were observed. Further, no substantial workload was detected in using the systems. However, the recommender system did create different user experiences across repeated sessions. Further, a diverse range of task loads is associated with the recommenders used in academia, from mixed designs with rich user controls to very few controls. This research provided us with insights that can be used to help designers incorporate and accommodate new features and take calculated risks when designing serendipitous education technology.

Influence of tweets and diversification on serendipitous research paper recommender systems.

In recent years, a large body of literature has accumulated around the topic of research paper recommender systems. However, since most studies have focused on the variable of accuracy, they have overlooked the serendipity of recommendations, which is an important determinant of user satisfaction. Serendipity is concerned with the relevance and unexpectedness of recommendations, and so serendipitous items are considered those which positively surprise users. The purpose of this article was to examine two key research questions: firstly, whether a user’s Tweets can assist in generating more serendipitous recommendations; and secondly, whether the diversification of a list of recommended items further improves serendipity. To investigate these issues, an online experiment was conducted in the domain of computer science with 22 subjects. As an evaluation metric, we use the serendipity score (SRDP), in which the unexpectedness of recommendations is inferred by using a primitive recommendation strategy. The results indicate that a user’s Tweets do not improve serendipity, but they can reflect recent research interests and are typically heterogeneous. Contrastingly, diversification was found to lead to a greater number of serendipitous research paper recommendations.

Triggers and connection-making for serendipity via user interface in recommender systems

This paper reports on the use of transparency in recommender a system that facilitates serendipitous encounters for users. Currently, there are serendipitous recommender systems that facilitate serendipitous encounters; however, there are no studies on the connection-making process or on the process of achieving connection-making through a user interface design. Adding to our previous work on connection-making and serendipity-facilitating recommender systems, we examine transparency in recommender systems as it relates to connection-making we studied transparency of recommendations to foster connection-making. This study is novel as it introduces a new user interface design for recommender system in academia and new study methods and approaches and studies a large group of users who are using this recommender system. The user interface components such as bubble messages on recommender system mechanism, user controls on manipulating the recommender system outcomes and showing authors work addition to recommendation. Repeated measure design of research was used to study serendipity and task load among users for Google Scholar and JabRef related work user interface (User interface developed for Experiment). Subjective evaluation of user interface was done along with NASA-Task Load Index for workload measurement. Further sentiment analysis was conducted for validations of findings. Our study finds that serendipitous recommendations and user satisfaction is facilitated via transparency in recommender systems. Furthermore, we found that transparency enhances interactivity for users who are looking for novel and useful recommendations related to their work. This work contributes to human computer interaction studies of recommender systems and reviews the leading literature on transparency, serendipity, and recommender systems in learning environments.

Holistic User Context-Aware Recommender Algorithm

Existing recommender algorithms lack dynamism, human focus, and serendipitous recommendations. The literature indicates that the context of a user influences user decisions, and when incorporated in recommender systems (RSs), novel and serendipitous recommendations can be realized. This article shows that social, cultural, psychological, and economic contexts of a user influence user traits or decisions. The article demonstrates a novel approach of incorporating holistic user context-aware knowledge in an algorithm to solve the highlighted problems. Web content mining and collaborative filtering approaches were used to develop a holistic user context-aware (HUC) algorithm. The algorithm was evaluated on a social network using online experimental evaluations. The algorithm demonstrated dynamism, novelty, and serendipity with an average of 84% novelty and 85% serendipity.

Facilitating research through serendipity of recommendations

Recommender systems are used to suggest items that are useful to users. The recommendations can be surprising and may be categorized as serendipitous recommendations. One of the limitations with serendipitous recommendations is that the user interface of such a recommender system rarely supports the user to switch from accuracy orientation to serendipity facilitations. Using serendipitous recommendations can be challenging. This is because the user might not fully benefit from and understand the serendipitous recommendations. One main advantage of this type of system is that a serendipity-oriented recommender system can be used for the supervision of research students. It can help them to find a novel topic in the area of their research interests. This paper reports on a novel user interface design for facilitating serendipitous recommendations generation in educational environments. The user interface of this recommender system provides students with user controls and visualization in order to explore research articles. This research comprises user experience experiments conducted in an academic environment and evaluated by means of a user centered design evaluation. It involves research articles recommender system named JabRef. The recommender systems were used by students at the undergraduate level. Users reported an enhanced user experience while using the user controls and visualization and serendipitous resource discovery. It was found that user interface design can facilitate a serendipity recommender system in the learning environment. University professors supervising students during the research can also benefit from the recommender system.

Transparency for Beyond-Accuracy Experiences: A Novel User Interface for Recommender Systems

Facilitating serendipitous recommendations via a user interface for a recommender system is challenging. Recommender systems that facilitate serendipitous recommendations create what are considered to be beyond-accuracy experiences. Because serendipitous recommendations are unexpected, users may experience some trust issues with the recommender system. This paper reports on the use of transparency for connection-making (a process/stage of serendipitous experience) in serendipity recommendations generation and examines trust with respect to an interactive serendipitous recommender system. We developed a user interface for JabRef (related work tab). The experiment studied the responses of users, comprised of undergraduate students, in a university environment. Data was collected for both subjective and objective evaluations. The analysis was used to determine the significance and impact of transparency in recommender systems in which serendipitous recommendations are generated. The results showed that greater transparency has a positive impact on user experience, helps with connection-making, and enhances the trust of learners who use recommender systems for learning purposes. Such systems can be used by faculty members and students during research and supervision in order to facilitate finding new research topics and trust in such systems.

User Control and Serendipitous Recommendations in Learning Environments

This paper reports about the study of serendipitous recommendations generation by recommender systems in educational environments. The recommender system provides students with serendipity slider to express desired serendipity and accuracy of recommendations. The recommender uses interface level processing of randomization of recommendations list. The increasing serendipity results in changing score/values of accuracy of recommendations. This was tested by deploying a study material recommender system based on collaborative filtering techniques. The recommender system was used by 60 students in a focus group setting. In conclusion, the research suggests that serendipitous recommendations can be archived using user controlled recommenders in learning environment.

Visualizing Serendipitous Recommendations in User Controlled Recommender System for Learning

In this paper, we report on a preliminary study about user preferences for recommender system visualizing serendipitous recommendations. A focus group study of fifty nine users (students) was studied for recording their preferences. The focus group was shown and explained the interaction with six(6) common recommender system visualization techniques. Multivariate analysis and LDA ( Linear Discriminant Analysis) and Clustering were performed to compute various visualization significance against various recommender systems attributes. The results showed there is difference in various types of recommender visualization when presenting/generating recommender results facilitating serendipity. This research enables software engineers and data scientist to design visualizations for recommender systems that focus users that need serendipitous recommendations presentation along with accuracy

Stakeholders Analysis for Serendipitous Recommenders system in Learning Environments

This paper reports on a 360 degree stakeholder analysis for the study and design of a study material recommender system at Institute of Management Sciences, Peshawar, Pakistan. During the survey we collected data from 300 students, 50 teachers and the librarian regarding the preferences towards study material recommender systems within the institute. Different than traditional study material recommender systems, we aim to provide a recommender system that enables the user to control on the outcomes of recommender system to accuracy until serendipitous recommendations. By doing so, we aim to empower the students and enable them to interact and control the provided recommendations in light of their respective needs. To further understand the user control needs of the stakeholders, we explicitly asked about user control aspects for the planned study material recommender system. We learned that there is strong need among all three stakeholder groups for user control of recommender systems.

Curumim: A Serendipitous Recommender System based on Human Curiosity

Tourism is an important source of income of countries, since nowadays 10% of GDP corresponds to a direct, indirect or induced effect of tourism. In European countries like Spain, the activity reached numbers like 11% of GDP in 2016. Therefore, providing an efficient and personalised service for tourists has become an essential issue in the development of new technological resources. This work aims to build a better experience for the tourist through the fusion of three axes: human psychology, namely curiosity, technological innovation and social networks. This article describes CURUMIM system, which, from data available on social networks, predicts the level of curiosity of a user and then, tied to other measures, generates novel and serendipitous recommendations of touristic places around the world. In other words, the recommendations will be accurate and adapted to the level of curiosity of a given user on one hand and, on the other, they will positively surprise the users.

A survey of serendipity in recommender systems

Recommender systems use past behaviors of users to suggest items. Most tend to offer items similar to the items that a target user has indicated as interesting. As a result, users become bored with obvious suggestions that they might have already discovered. To improve user satisfaction, recommender systems should offer serendipitous suggestions: items not only relevant and novel to the target user, but also significantly different from the items that the user has rated. However, the concept of serendipity is very subjective and serendipitous encounters are very rare in real-world scenarios, which makes serendipitous recommendations extremely difficult to study. To date, various definitions and evaluation metrics to measure serendipity have been proposed, and there is no wide consensus on which definition and evaluation metric to use. In this paper, we summarize most important approaches to serendipity in recommender systems, compare different definitions and formalizations of the concept, discuss serendipity-oriented recommendation algorithms and evaluation strategies to assess the algorithms, and provide future research directions based on the reviewed literature.