The closed loop between opinion formation and personalised recommendations
University College Groningen, University of Groningen（格罗宁根大学格罗宁根大学学院）；University of Twente（特文特大学）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；CNRS（法国国家科学研究中心）；Inria（法国国家信息与自动化研究所）；Grenoble INP（格勒诺布尔国立高等物理、数学与信息学院）
Wilbert Samuel Rossi, Jan Willem Polderman, Paolo Frasca
In online platforms, recommender systems are responsible for directing users to relevant contents. In order to enhance the users' engagement, recommender systems adapt their output to the reactions of the users, who are in turn affected by the recommended contents. In this work, we study a tractable analytical model of a user that interacts with an online news aggregator, with the purpose of making explicit the feedback loop between the evolution of the user's opinion and the personalised recommendation of contents. More specifically, we assume that the user is endowed with a scalar opinion about a certain issue and seeks news about it on a news aggregator: this opinion is influenced by all received news, which are characterized by a binary position on the issue at hand. The user is affected by a confirmation bias, that is, a preference for news that confirm her current opinion. The news aggregator recommends items with the goal of maximizing the number of user's clicks (as a measure of her engagement): in order to fulfil its goal, the recommender has to compromise between exploring the user's preferences and exploiting what it has learned so far. After defining suitable metrics for the effectiveness of the recommender systems (such as the click-through rate) and for its impact on the opinion, we perform both extensive numerical simulations and a mathematical analysis of the model. We find that personalised recommendations markedly affect the evolution of opinions and favor the emergence of more extreme ones: the intensity of these effects is inherently related to the effectiveness of the recommender. We also show that by tuning the amount of randomness in the recommendation algorithm, one can seek a balance between the effectiveness of the recommendation system and its impact on the opinions.
Point-of-Interest (POI) recommender systems play a vital role in people's lives by recommending unexplored POIs to users and have drawn extensive attention from both academia and industry. Despite their value, however, they still suffer from the challenges of capturing complicated user preferences and fine-grained user-POI relationship for spatio-temporal sensitive POI recommendation. Existing recommendation algorithms, including both shallow and deep approaches, usually embed the visiting records of a user into a single latent vector to model user preferences: this has limited power of repres…
Recommendation systems personalise suggestions to individuals to help them in their decision making and exploration tasks. In the ideal case, these recommendations, besides of being accurate, should also be novel and explainable. However, up to now most platforms fail to provide both, novel recommendations that advance users' exploration along with explanations to make their reasoning more transparent to them. For instance, a well-known recommendation algorithm, such as matrix factorisation (MF), optimises only the accuracy criterion, while disregarding other quality criteria such as the expla…
Algorithmic fairness for artificial intelligence has become increasingly relevant as these systems become more pervasive in society. One realm of AI, recommender systems, presents unique challenges for fairness due to trade offs between optimizing accuracy for users and fairness to providers. But what is fair in the context of recommendation--particularly when there are multiple stakeholders? In an initial exploration of this problem, we ask users what their ideas of fair treatment in recommendation might be, and why. We analyze what might cause discrepancies or changes between user's opinions…
Recommender systems are a class of machine learning algorithms that provide relevant recommendations to a user based on the user's interaction with similar items or based on the content of the item. In settings where the content of the item is to be preserved, a content-based approach would be beneficial. This paper aims to highlight the advantages of the content-based approach through learned embeddings and leveraging these advantages to provide better and personalised movie recommendations based on user preferences to various movie features such as genre and keyword tags.
Recommendation systems have received considerable attention recently. However, most research has been focused on improving the performance of collaborative filtering (CF) techniques. Social networks, indispensably, provide us extra information on people's preferences, and should be considered and deployed to improve the quality of recommendations. In this paper, we propose two recommendation models, for individuals and for groups respectively, based on social contagion and social influence network theory. In the recommendation model for individuals, we improve the result of collaborative filte…
Nova School of Business and Economics, Universidade NOVA de Lisboa（新里斯本大学商业与经济学院）
In contrast to many other domains, recommender systems in health services may benefit particularly from the incorporation of health domain knowledge, as it helps to provide meaningful and personalised recommendations catering to the individual's health needs. With recent advances in representation learning enabling the hierarchical embedding of health knowledge into the hyperbolic Poincare space, this work proposes a content-based recommender system for patient-doctor matchmaking in primary care based on patients' health profiles, enriched by pre-trained Poincare embeddings of the ICD-9 codes …
Indian Institute of Technology Kharagpur（印度理工学院克勒格布尔分校）
Recommender systems are one of the most widely used services on several online platforms to suggest potential items to the end-users. These services often use different machine learning techniques for which fairness is a concerning factor, especially when the downstream services have the ability to cause social ramifications. Thus, focusing on the non-personalised (global) recommendations in news media platforms (e.g., top-k trending topics on Twitter, top-k news on a news platform, etc.), we discuss on two specific fairness concerns together (traditionally studied separately)---user fairness …
Bangladesh University of Engineering and Technology (BUET)（孟加拉国工程技术大学（BUET））
Location-based Social Networks (LBSNs) enable users to socialize with friends and acquaintances by sharing their check-ins, opinions, photos, and reviews. Huge volume of data generated from LBSNs opens up a new avenue of research that gives birth to a new sub-field of recommendation systems, known as Point-of-Interest (POI) recommendation. A POI recommendation technique essentially exploits users' historical check-ins and other multi-modal information such as POI attributes and friendship network, to recommend the next set of POIs suitable for a user. A plethora of earlier works focused on tra…
International Institute of Information Technology（国际信息技术学院）；University of California, San Diego（加州大学圣地亚哥分校）
We investigate a growing body of work that seeks to improve recommender systems through the use of review text. Generally, these papers argue that since reviews 'explain' users' opinions, they ought to be useful to infer the underlying dimensions that predict ratings or purchases. Schemes to incorporate reviews range from simple regularizers to neural network approaches. Our initial findings reveal several discrepancies in reported results, partly due to (e.g.) copying results across papers despite changes in experimental settings or data pre-processing. First, we attempt a comprehensive analy…
Universidad Autónoma de Madrid（马德里自治大学）；Pontificia Universidad Católica de Chile（智利天主教大学）
The success of neural network embeddings has entailed a renewed interest in using knowledge graphs for a wide variety of machine learning and information retrieval tasks. In particular, current recommendation methods based on graph embeddings have shown state-of-the-art performance. These methods commonly encode latent rating patterns and content features. Different from previous work, in this paper, we propose to exploit embeddings extracted from graphs that combine information from ratings and aspect-based opinions expressed in textual reviews. We then adapt and evaluate state-of-the-art gra…
Cold-start is a very common and still open problem in the Recommender Systems literature. Since cold start items do not have any interaction, collaborative algorithms are not applicable. One of the main strategies is to use pure or hybrid content-based approaches, which usually yield to lower recommendation quality than collaborative ones. Some techniques to optimize performance of this type of approaches have been studied in recent past. One of them is called feature weighting, which assigns to every feature a real value, called weight, that estimates its importance. Statistical techniques fo…
Before A/B testing online a new version of a recommender system, it is usual to perform some offline evaluations on historical data. We focus on evaluation methods that compute an estimator of the potential uplift in revenue that could generate this new technology. It helps to iterate faster and to avoid losing money by detecting poor policies. These estimators are known as counterfactual or off-policy estimators. We show that traditional counterfactual estimators such as capped importance sampling and normalised importance sampling are experimentally not having satisfying bias-variance compro…
Over the last few years, the arena of mobile application development has expanded considerably beyond the balance of the worldś software markets. With the growing number of mobile software companies, and the mounting sophistication of smartphones\' technology, developers have been building several categories of applications on dissimilar platforms. However, developers confront several challenges through the implementation of mobile application projects. In particular, there is a lack of consolidated systems that are competent to provide developers with personalised services promptly and effici…
Aggregated data in real world recommender applications often feature fat-tailed distributions of the number of times individual items have been rated or favored. We propose a model to simulate such data. The model is mainly based on social interactions and opinion formation taking place on a complex network with a given topology. A threshold mechanism is used to govern the decision making process that determines whether a user is or is not interested in an item. We demonstrate the validity of the model by fitting attendance distributions from different real data sets. The model is mathematical…
We present a framework to generate and evaluate thematic recommendations based on multilayer network representations of knowledge graphs (KGs). In this representation, each layer encodes a different type of relationship in the KG, and directed interlayer couplings connect the same entity in different roles. The relative importance of different types of connections is captured by an intuitive salience matrix that can be estimated from data, tuned to incorporate domain knowledge, address different use cases, or respect business logic. We apply an adaptation of the personalised PageRank algorithm…
University of Westminster（威斯敏斯特大学）；Informatics Institute of Technology（信息技术学院）
Recommendations Systems allow users to identify trending items among a community while being timely and relevant to the user's expectations. When the purpose of various Recommendation Systems differs, the required type of recommendations also differs for each use case. While one Recommendation System may focus on recommending popular items, another may focus on recommending items that are comparable to the user's interests. Content-based filtering, user-to-user & item-to-item Collaborative filtering, and more recently; Deep Learning methods have been brought forward by the researchers to achie…
The identification of strategic business partnerships can potentially provide competitive advantages for businesses; however, due to the dynamics and uncertainty present in business environments, this task could be challenging. To help businesses in this task, this study presents a similarity model between businesses that consider the opinions of users on content shared by businesses on social media. Thus, this model captures significant virtual relationships among businesses that are generated by users in the virtual world. Besides, we propose an algorithm for detecting business communities i…
Recommender systems are important to help users select relevant and personalised information over massive amounts of data available. We propose an unified framework called Preference Network (PN) that jointly models various types of domain knowledge for the task of recommendation. The PN is a probabilistic model that systematically combines both content-based filtering and collaborative filtering into a single conditional Markov random field. Once estimated, it serves as a probabilistic database that supports various useful queries such as rating prediction and top-$N$ recommendation. To handl…
We consider a discrete opinion formation problem in a setting where agents are influenced by both information diffused by their social relations and from recommendations received directly from the social media manager. We study how the "strength" of the influence of the social media and the homophily ratio affect the probability of the agents of reaching a consensus and how these factors can determine the type of consensus reached. In a simple 2-symmetric block model we prove that agents converge either to a consensus or to a persistent disagreement. In particular, we show that when the homoph…
USC Information Sciences Institute（南加州大学信息科学研究所）；University of Maryland（马里兰大学）
Social media users have finite attention which limits the number of incoming messages from friends they can process. Moreover, they pay more attention to opinions and recommendations of some friends more than others. In this paper, we propose LA-LDA, a latent topic model which incorporates limited, non-uniformly divided attention in the diffusion process by which opinions and information spread on the social network. We show that our proposed model is able to learn more accurate user models from users' social network and item adoption behavior than models which do not take limited attention in…
Existing socio-psychological studies suggest that users of a social network form their opinions relying on the opinions of their neighbors. According to DeGroot opinion formation model, one value of particular importance is the asymptotic consensus value---the sum of user opinions weighted by the users' eigenvector centralities. This value plays the role of an attractor for the opinions in the network and is a lucrative target for external influence. However, since any potentially malicious control of the opinion distribution in a social network is clearly undesirable, it is important to desig…
Nowadays, recommender systems are present in many daily activities such as online shopping, browsing social networks, etc. Given the rising demand for reinvigoration of the tourist industry through information technology, recommenders have been included into tourism websites such as Expedia, Booking or Tripadvisor, among others. Furthermore, the amount of scientific papers related to recommender systems for tourism is on solid and continuous growth since 2004. Much of this growth is due to social networks that, besides to offer researchers the possibility of using a great mass of available and…
City University of Hong Kong（香港城市大学）；Michigan State University（密歇根州立大学）；The Hong Kong Polytechnic University（香港理工大学）
In recent years, Graph Neural Networks (GNNs), which can naturally integrate node information and topological structure, have been demonstrated to be powerful in learning on graph data. These advantages of GNNs provide great potential to advance social recommendation since data in social recommender systems can be represented as user-user social graph and user-item graph; and learning latent factors of users and items is the key. However, building social recommender systems based on GNNs faces challenges. For example, the user-item graph encodes both interactions and their associated opinions;…
We have conducted three empirical studies of the effects of friend recommendations and general ratings on how online users make choices. These two components of social influence were investigated through user studies on Mechanical Turk. We find that for a user deciding between two choices an additional rating star has a much larger effect than an additional friend's recommendation on the probability of selecting an item. Equally important, negative opinions from friends are more influential than positive opinions, and people exhibit more random behavior in their choices when the decision invol…
University of Westminster（威斯敏斯特大学）；Informatics Institute of Technology（信息技术学院）
Recommendations Systems have been identified to be one of the integral elements of driving sales in e-commerce sites. The utilization of opinion mining data extracted from trends has been attempted to improve the recommendations that can be provided by baseline methods in this research when user-click data is lacking or is difficult to be collected due to privacy concerns. Utilizing social trends to influence the recommendations generated for a set of unique items has been explored with the use of a suggested scoring mechanism. Embracing concepts from decentralized networks that are expected t…
We present a user of model interaction based on the physics of kinetic exchange, and extend it to individuals placed in a grid with local interaction. We show with numerical analysis and partial analytical results that the critical symmetry breaking transitions and percolation effects typical of the full interaction model do not take place if the range of interaction is limited, allowing for the co-existence of majorty and minority opinions in the same community. We then introduce a peer recommender system in the model, showing that, even with very local iteraction and a small probability of a…
The effects of social media on critical issues, such as polarization and misinformation, are under scrutiny due to the disruptive consequences that these phenomena can have on our societies. Among the algorithms routinely used by social media platforms, people-recommender systems are of special interest, as they directly contribute to the evolution of the social network structure, affecting the information and the opinions users are exposed to. In this paper, we propose a framework to assess the effect of people recommenders on the evolution of opinions. Our proposal is based on Monte Carlo si…
In the Internet era, online social media emerged as the main tool for sharing opinions and information among individuals. In this work we study an adaptive model of a social network where directed links connect users with similar tastes, and over which information propagates through social recommendation. Agent-based simulations of two different artificial settings for modeling user tastes are compared with patterns seen in real data, suggesting that users differing in their scope of interests is a more realistic assumption than users differing only in their particular interests. We further in…
Recommender systems are crucial to alleviate the information overload problem in online worlds. Most of the modern recommender systems capture users' preference towards items via their interactions based on collaborative filtering techniques. In addition to the user-item interactions, social networks can also provide useful information to understand users' preference as suggested by the social theories such as homophily and influence. Recently, deep neural networks have been utilized for social recommendations, which facilitate both the user-item interactions and the social network information…
Peer recommendation is a crowdsourcing task that leverages the opinions of many to identify interesting content online, such as news, images, or videos. Peer recommendation applications often use social signals, e.g., the number of prior recommendations, to guide people to the more interesting content. How people react to social signals, in combination with content quality and its presentation order, determines the outcomes of peer recommendation, i.e., item popularity. Using Amazon Mechanical Turk, we experimentally measure the effects of social signals in peer recommendation. Specifically, a…
Nowadays, we live in a society where people often form their opinion by accessing and discussing contents shared on social networking websites. While these platforms have fostered information access and diffusion, they represent optimal environments for the proliferation of polluted contents, which is argued to be one of the co-causes of polarization/radicalization. Moreover, recommendation algorithms - intended to enhance platform usage - are likely to augment such phenomena, generating the so called Algorithmic Bias. In this work, we study the impact that different network topologies have on…
An Interpretable Recommendation Model for Psychometric Data, With an Application to Gerontological Primary Care
Institute of Mathematics；Computer Science, University of S\ ao Paulo S\ ao Carlos Brazil；Department of Gerontology, S\ ao Carlos Federal University S\ ao Carlos Brazil；School of Arts, Sciences；Humanities, University of S\ ao Paulo S\ ao Paulo Brazil；Computer Science, University of S\ ao Paulo；Department of Gerontology, S\ ao Carlos Federal University；Humanities, University of S\ ao Paulo
How to make the best decision between the opinions and tastes of your friends and acquaintances? Therefore, recommender systems are used to solve such issues. The common algorithms use a similarity measure to predict active users' tastes over a particular item. According to the cold start and data sparsity problems, these systems cannot predict and suggest particular items to users. In this paper, we introduce a new recommender system is able to find user preferences and based on it, provides the recommendations. Our proposed system called CUPCF is a combination of two similarity measures in c…
Social network websites, such as Facebook, YouTube, Lastfm etc, have become a popular platform for users to connect with each other and share content or opinions. They provide rich information for us to study the influence of user's social circle in their decision process. In this paper, we are interested in examining the effectiveness of social network information to predict the user's ratings of items. We propose a novel hierarchical Bayesian model which jointly incorporates topic modeling and probabilistic matrix factorization of social networks. A major advantage of our model is to automat…
