Since the emergence of Recommender Systems (RS), most of the research has focused on improving the capability of a recommender system to predict and provide an accurate recommendation. However, the literature has demonstrated increasing evidence that providing accurate recommendations is not sufficient to increase users’ acceptance of the provided recommendations. Hence, it is vital for a recommender system to focus not only on the accuracy of the provided recommendations but also on other factors that influence the acceptance of recommendations and the extent to which these recommendations are convincing or persuasive. Consequently, there becomes a need for new research paradigms to help improve the capabilities of recommender systems, which goes beyond the recommendation accuracy. One of the recently emerged research directions that consider this need fosters the idea of adopting human-related theories from the social sciences domain, such as persuasiveness of social communication. In this context, however, a challenging, non-trivial, and not fully explored issue that arises is: how to integrate human-related theories into a recommender system to be one of its intrinsic characteristics in order to improve its performance beyond its accuracy? This thesis aims to address the above issue from two angles: first, it investigates improving recommender systems by increasing users’ acceptance of the recommendations. To achieve this, the influence of persuasion principles on users of recommender systems is investigated. Then a reference architecture framework to adapt and integrate persuasion features as a substantial characteristic of recommender systems is proposed. The proposed framework, named Personalized Persuasive RS (PerPer), adopts concepts from the social sciences literature, namely personality traits and persuasion principles. In addition, PerPer adapts machine learning concepts, in particular, the Learning Automata, to support its learning capabilities. Second, the thesis discusses evaluating recommender systems beyond their accuracy. Particularly, it proposes two evaluation approaches that aim to evaluate recommender systems in a comprehensive way that goes beyond evaluating accuracy only. The first evaluation approach is called the Comprehensive Performance evaluation (ComPer). It adopts concepts from the human learning domain and provides a simple, thorough, and setting-independent evaluation approach for recommenders. The essence of ComPer is to consider a recommender system as a human being, and hence the former’s outcomes (i.e., recommendations) can be evaluated and validated in a way similar to how humans’ learning outcomes are evaluated. The second evaluation approach adopts goal-oriented modeling to provide an evaluation that does not only assess recommenders beyond their accuracy but also considers the multi-stakeholders of RSs. We demonstrate, empirically, and by user studies, the feasibility and usefulness of the proposed approaches. The contributions of the thesis are: (1) A characterization of recommender systems as systems supported with human traits and features, which goes beyond the conventional recommender systems known in the literature. (2) A user study that examines the impact of persuasive principles on users of recommender systems. (3) A Personalized Persuasive RS (PerPer) reference architecture framework to enrich recommender systems with persuasion capabilities that are personalized and adaptive for different users. (4) A mapping between human’s cognitive skills and the recommendation process. (5) The Comprehensive Performance evaluation (ComPer) framework to provide a comprehensive assessment of recommender systems considering multiple evaluation dimensions other than accuracy. And (6) a goal-oriented evaluation approach to assess the impact of multiple alternatives for recommendation approaches on the satisfaction of RSs stakeholders’ goals.

An individual’s ability to locate the information they desire grows more slowly than the rate at which new information becomes available. Customers are constantly confronted with situations in which they have many options to choose from and need assistance exploring or narrowing down the possibilities. Recommender systems are one tool to help bridge this gap. There are various mechanisms being employed to create recommender systems, but the most common systems fall into two main classes: content-based and collaborative filtering systems. Content-based recommender systems match the textual information of a particular product with the textual information representing the interests of a customer. Collaborative filtering systems use patterns in customer ratings to make recommendations. Both types of recommender systems require significant data resources in the form of a customer’s ratings and product features; hence they are not able to generate high quality recommendations. Hybrid mechanisms have been used by researchers to improve the performance of recommender systems where one can integrate more than one mechanism to overcome the drawbacks of an individual system. The hybrid approach proposed in this thesis is the integration of content and context-based with collaborative filtering, since these are the most successful and widely used mechanisms. This proposed approach will look into the integration of content and context data with rating data using a different mechanism that mainly focuses on boosting a customer’s trust in the recommender system. Researchers have been trying to improve system performance using hybrid approaches, but research is lacking on providing justifications for recommended products. Hence, the proposed approach will mainly focus on providing justifications for recommended products as this plays a crucial role in obtaining the satisfaction and trust of customers. A product’s features and a customer’s context attributes are used to provide justifications. In addition to this, the presentation mechanism needs to be very effective as it has been observed that customers trust more in a system when there are explanations on how the recommended products have been computed and presented. Finally, this proposed recommender system will allow the customer to interact with it in various ways to provide feedback on the recommendations and justifications. Overall, this integration will be very useful in achieving a stronger correlation between the customers and products. Experimental results clearly showed that the majority of the participants prefer to have recommendations with their justifications and they received valuable recommendations on which they could trust.

Machine learning algorithms are increasingly used by decision making systems that affect individual lives in a wide variety of ways. Consequently, in recent years concerns have been raised about the social and ethical implications of using such algorithms. Particular concerns include issues surrounding privacy, fairness, and transparency in decision systems. This dissertation introduces new tools and measures for improving the social desirability of data-driven decision systems, and consists of two main parts. The first part provides a useful tool for an important class of decision making algorithms: collaborative filtering in recommender systems. In particular, it introduces the idea of improving socially relevant properties of a recommender system by augmenting the input with additional training data, an approach which is inspired by prior work on data poisoning attacks and adapts them to generate `antidote data' for social good. We provide an algorithmic framework for this strategy and show that it can efficiently improve the polarization and fairness metrics of factorization-based recommender systems. In the second part, we focus on fairness notions that incorporate data inputs used by decision systems. In particular, we draw attention to `data minimization', an existing principle in data protection regulations that restricts a system to use the minimal information that is necessary for performing the task at hand. First, we propose an operationalization for this principle that is based on classification accuracy, and we show how a natural dependence of accuracy on data inputs can be expressed as a trade-off between fair-inputs and fair-outputs. Next, we address the problem of auditing black- box prediction models for data minimization compliance. For this problem, we suggest a metric for data minimization that is based on model instability under simple imputations, and we extend its applicability from a finite sample model to a distributional setting by introducing a probabilistic data minimization guarantee. Finally, assuming limited system queries, we formulate the problem of allocating a query budget to simple imputations for investigating model instability as a multi-armed bandit framework, for which we design efficient exploration strategies.

Indiana University-Purdue University Indianapolis (IUPUI) / As the amount of information grows, the desire to efficiently filter out unnecessary information and retain relevant or interested information for people is increasing. To extract the information that will be of interest to people efficiently, we can utilize recommender systems. Recommender systems are information filtering systems that predict the preference of a user to an item. Based on historical data of users, recommender systems are able to make relevant recommendations to users. Due to its usefulness, Recommender systems have been widely used in many applications, including e-commerce and healthcare information systems. However, existing recommender systems suffer from several issues, including data sparsity and user/item heterogeneity. In this thesis, a hybrid dynamic and multi-collaborative filtering based recommendation technique has been developed to recommend search terms for physicians when physicians review a large number of patients’ information. Besides, a local sparse linear method ensemble has been developed to tackle the issues of data sparsity and user/item heterogeneity. In health information technology systems, most physicians suffer from information overload when they review patient information. A novel hybrid dynamic and multi-collaborative filtering method has been developed to improve information retrieval from electronic health records. We tackle the problem of recommending the next search term to a physician while the physician is searching for information about a patient. In this method, I have combined first-order Markov Chain and multi-collaborative filtering methods. For multi-collaborative filtering methods, I have developed the physician-patient collaborative filtering and transition-involved collaborative filtering methods. The developed method is tested using electronic health record data from the Indiana Network for Patient Care. The experimental results demonstrate that for 46.7% of test cases, this new method is able to correctly prioritize relevant information among top-5 recommendations that physicians are truly interested in. The local sparse linear model ensemble has been developed to tackle both the data sparsity and the user/item heterogeneity issues for the top-n recommendation. Multiple local sparse linear models are learned for all the users and items in the system. I have developed similarity-based and popularity-based methods to determine the local training data for each local model. Each local model is trained on Sparse Linear Method (SLIM) which is a powerful recommendation technique for top-n recommendation. These learned models are then combined in various ways to produce top-N recommendations. I have developed model results combination and model combination methods to combine all learned local models. The developed methods are tested on a benchmark dataset and its sparsified datasets. The experiments demonstrate 18.4% improvement from such ensemble models, particularly on sparse datasets.

Indiana University-Purdue University Indianapolis (IUPUI) / Over the past decade, universities had a significant improvement in using online learning tools. A standard learning management system provides fundamental functionalities to satisfy the basic needs of its users. The new generation of learning management systems have introduced a novel system that provides social networking features. An unprecedented number of users use the social aspects of such platforms to create their profile, collaborate with other users, and find their desired career path. Nowadays there are many learning systems which provide learning materials, certificates, and course management systems. This allows us to utilize such information to help the students and the instructors in their academic life. The presented research work's primary goal is to focus on creating an intelligent personal assistant within the social learning systems. The proposed personal assistant has a human-like persona, learns about the users, and recommends useful and meaningful materials for them. The designed system offers a set of features for both institutions and members to achieve their goal within the learning system. It recommends jobs and friends for the users based on their profile. The proposed agent also prioritizes the messages and shows the most important message to the user. The developed software supports model-controller-view architecture and provides a set of RESTful APIs which allows the institutions to integrate the proposed intelligent agent with their learning system.

This thesis aims to explore how we can empower users through using recommendation systems. Recommendation systems are all around us on the web, however today’s implementation is not necessarily that interactive. Literature argues for interaction designers to learn and get involved in the design of recommender systems and processes where AI technology is implemented. Moreover, designing with Artificial Intelligence as a material is explored and what interaction designers need to keep in mind when designing with it. The design process sets out to explore new ways of interacting with recommendation systems on the web. The process has used various design activities which generated several design proposals. These design proposals have been put through user testing using anonymous participants, in order to find a final design solution. The main conclusions are that informing and including users how the system works are important factors if we want users to interact with the recommender systems. It is not necessarily the most interactive design proposal that is the best solution, but rather one that is likely to interact with.

More and more companies are making online presence by opening online stores and providing customers with company and products information but the overwhelming amount of information also creates information overload for the customers. Customers feel frustrated when given too many choices while companies face the problem of turning browsers into actual buyers. Online recommender systems have been adopted to facilitate customer product search and provide personalized recommendation in the market place. The study will compare the persuasiveness of different online recommender systems and the factors influencing customer preferences. Review of the literature does show that online recommender systems provide customers with more choices, less effort, and better accuracy. Recommender systems using different technologies have been compared for their accuracy and effectiveness. Studies have also compared online recommender systems with human recommendations 4 and recommendations from expert systems. The focus of the comparison in this study is on the recommender systems using different methods to solicit product preference and develop recommendation message. Different from the technology adoption and acceptance models, the persuasive theory used in the study is a new perspective to look at the end user issues in information systems. This study will also evaluate the impact of product complexity and product involvement on recommendation persuasiveness. The goal of the research is to explore whether there are differences in the persuasiveness of recommendation given by different recommender systems as well as the underlying reasons for the differences. Results of this research may help online store designers and ecommerce participants in selecting online recommender systems so as to improve their products target and advertisement efficiency and effectiveness.

Learning on relational data is a relevant task in the machine learning community. Extracting information from structured data is a non-trivial task due to the combinatorial complexity of the domain and the necessity to construct methods that work on collections of values of different sizes rather than fixed representations. Relational data can naturally be interpreted as graphs, a class of flexible and expressive structures that can model data from diverse domains,from biology to social interactions. Graphs have been used in a huge variety of contexts, such as molecular modelling, social networks, image processing and recommendation systems. In this manuscript, we tackle some challenges in learning on relational data by developing new learning methodologies. Specifically, in our first contribution, we introduce a new class of metrics for relational data based on relational features extraction technique called Type ExtensionTrees. This class of metrics defines the (dis)similarity of two nodes in a graph by exploiting the nested structure of their relational neighborhood at different depth steps. In our second contribution, we developed a new strategy to collect the information of multisets of data values by introducing a new framework of learnable aggregators called Learning Aggregation Functions.We provide a detailed description of the methodologies and an extensive experimental evaluation on synthetic and real world data to assess the expressiveness of the proposed models. A particular focus is given to the application of these methods to the recommendation systems domain, exploring the combination of the proposed methods with recent techniques developed for Constructive Preference Elicitation and Group Recommendation tasks.

This thesis provides a novel approach to using data mining for e-commerce. The focus of our work is to apply association rule mining to collaborative recommender systems, which recommend articles to a user on the basis of other users' ratings for these articles as well as the similarities between this user's and other users' tastes. In this work, we propose a new algorithm for association rule mining specially tailored for use in collaborative recommendation. We make recommendations by exploring associations between users, associations between articles, and a combination of the two. We experimentally evaluated our approach on real data for many different parameter settings and compared its performance with that of other approaches under similar experimental conditions. Through our analysis and experiments, we have found that association rules are quite appropriate for collaborative recommendation domains and that they can achieve a performance that is comparable to current state of the art in recommender systems research.

Traditional hybrid recommender systems typically follow a manually created fixed prediction strategy in their decision making process. Experts usually design these static strategies as fixed combinations of different techniques. However, people&#039 / s tastes and desires are temporary and they gradually evolve. Moreover, each domain has unique characteristics, trends and unique user interests. Recent research has mostly focused on static hybridization schemes which do not change at runtime. In this thesis work, we describe an adaptive hybrid recommender system, called AdaRec that modifies its attached prediction strategy at runtime according to the performance of prediction techniques (user feedbacks). Our approach to this problem is to use adaptive prediction strategies. Experiment results with datasets show that our system outperforms naive hybrid recommender.