Review personalization aims at presenting the most relevant reviews of a product according to the preferences of the individual user. Existing studies of review personalization use the reviews authored by the user as a proxy for their preferences, and henceforth as a means for learning and evaluating personalization quality. In this work, we suggest using review votes rather than authorship for personalization. We propose MAGLLM, an approach that leverages heterogeneous graphs for modeling the relationships among reviews, products, and users, with large language model (LLM) to enrich user representation on the graph. Our evaluation over a unique public dataset that includes user voting information indicates that the vote signal yields substantially higher personalization performance across a variety of recommendation methods and e-commerce domains. It also indicates that our graph-LLM approach outperforms comparative baselines and algorithmic alternatives. We conclude with concrete recommendations for e-commerce platforms seeking to enhance their review personalization experience.

Category trees, or taxonomies, are rooted trees where each node, called a category, corresponds to a set of related items. The construction of taxonomies has been studied in various domains, including e-commerce, document management, and question answering. Multiple algorithms for automating construction have been proposed, employing a variety of clustering approaches and crowdsourcing. However, no formal model to capture such categorization problems has been devised, and their complexity has not been studied. To address this, we propose in this work a combinatorial model that captures many practical settings and show that the aforementioned empirical approach has been warranted, as we prove strong inapproximability bounds for various problem variants and special cases when the goal is to produce a categorization of the maximum utility. In our model, the input is a set of n weighted item sets that the tree would ideally contain as categories. Each category, rather than perfectly match the corresponding input set, is allowed to exceed a given threshold for a given similarity function. The goal is to produce a tree that maximizes the total weight of the sets for which it contains a matching category. A key parameter is an upper bound on the number of categories an item may belong to, which produces the hardness of the problem, as initially each item may be contained in an arbitrary number of input sets. For this model, we prove inapproximability bounds, of order \(\tilde{\Theta }(\sqrt {n})\) or \(\tilde{\Theta }(n)\) , for various problem variants and special cases, loosely justifying the aforementioned heuristic approach. Our work includes reductions based on parameterized randomized constructions that highlight how various problem parameters and properties of the input may affect the hardness. Moreover, for the special case where the category must be identical to the corresponding input set, we devise an algorithm whose approximation guarantee depends solely on a more granular parameter, allowing improved worst-case guarantees, as well as the application of practical exact solvers. We further provide efficient algorithms with much improved approximation guarantees for practical special cases where the cardinalities of the input sets or the number of input sets each items belongs to are not too large. Finally, we also generalize our results to DAG-based and non-hierarchical categorization.

Category trees, or taxonomies, are rooted trees where each node, called a category, corresponds to a set of related items. The construction of taxonomies has been studied in various domains, including e-commerce, document management, and question answering. Multiple algorithms for automating construction have been proposed, employing a variety of clustering approaches and crowdsourcing. However, no formal model to capture such categorization problems has been devised, and their complexity has not been studied. To address this, we propose in this work a combinatorial model that captures many practical settings and show that the aforementioned empirical approach has been warranted, as we prove strong inapproximability bounds for various problem variants and special cases when the goal is to produce a categorization of the maximum utility. In our model, the input is a set of n weighted item sets that the tree would ideally contain as categories. Each category, rather than perfectly match the corresponding input set, is allowed to exceed a given threshold for a given similarity function. The goal is to produce a tree that maximizes the total weight of the sets for which it contains a matching category. A key parameter is an upper bound on the number of categories an item may belong to, which produces the hardness of the problem, as initially each item may be contained in an arbitrary number of input sets. For this model, we prove inapproximability bounds, of order Θ̃(√n) or Θ̃(n), for various problem variants and special cases, loosely justifying the aforementioned heuristic approach. Our work includes reductions based on parameterized randomized constructions that highlight how various problem parameters and properties of the input may affect the hardness. Moreover, for the special case where the category must be identical to the corresponding input set, we devise an algorithm whose approximation guarantee depends solely on a more granular parameter, allowing improved worst-case guarantees. Finally, we also generalize our results to DAG-based and non-hierarchical categorization.

Category trees play a central role in many web applications, enabling browsing-style information access. Building trees that reflect users' dynamic interests is, however, a challenging task, carried out by taxonomists. This manual construction leads to outdated trees as it is hard to keep track of market trends. While taxonomists can identify candidate categories, i.e. sets of items with a shared label, most such categories cannot simultaneously exist in the tree, as platforms set a bound on the number of categories an item may belong to. To address this setting, we formalize the problem of constructing a tree where the categories are maximally similar to desirable candidate categories while satisfying combinatorial requirements and provide a model that captures practical considerations.

Search mechanisms over large assortments of items are central to the operation of many platforms. As users commonly express filtering conditions based on item properties that are not initially stored, companies must derive the missing information by training and applying binary classifiers. Choosing which classifiers to construct is however not trivial, since classifiers differ in construction costs and range of applicability. Previous work has considered the problem of selecting a classifier set of minimum construction cost, but this has been done under the (often unrealistic) assumption that the available budget is unlimited and allows to support all search queries. In practice, budget constraints require prioritizing some queries over others. To capture this consideration, we study in this work a more general model that allows assigning to each search query a score that models how important it is to compute its result set and examine the optimization problem of selecting a classifier set, whose cost is within the budget, that maximizes the overall score of the queries it can answer. We show that this generalization is likely much harder to approximate complexity-wise, even assuming limited special cases. Nevertheless, we devise a heuristic algorithm, whose effectiveness is demonstrated in our experimental study over real-world data, consisting of a public dataset and datasets provided by a large e-commerce company that include costs and scores derived by business analysts. Finally, we show that our methods are applicable also for related problems in practical settings where there is some flexibility in determining the budget.

Our ability to collect data is rapidly outstripping our ability to effectively store and use it. Organizations are therefore facing tough decisions of what data to archive (or dispose of) to effectively meet their business goals. PHOcus addresses this problem in the context of image data (photos) by proposing which photos to archive to meet an online storage budget. The decision is based on factors such as usage patterns and their relative importance, the quality and size of a photo, the relevance of a photo for a usage pattern, the similarity between different photos, as well as policy requirements of what photos must be retained. We formalize the photo archival problem and give an efficient algorithm with an optimal approximation guarantee. We then demonstrate our system, PHOcus, on an e-commerce application as well as with personal photos on a smartphone, and discuss how many of the inputs to the problem can be automatically obtained.

Category trees play a central role in e-commerce platforms, enabling browsing-style information access. Building category trees that reflect users' dynamic information needs is a challenging task, mostly carried out by in-house taxonomists. This manual construction often leads to trees that are lacking or outdated since it is hard to keep track of market trends, seasonal changes, holidays, and special events.To support a browsing experience that better matches the user information needs, and to considerably reduce the manual work performed by taxonomists, we propose CONCAT - a system that leverages the demand-based nature of the query paradigm to automatically build a category tree that is maximally similar to the result sets for search queries. We demonstrate the effectiveness of CONCAT on real-world data, taken from a large e-commerce platform, by interacting with the ICDE'21 participants who act both as the consumers and the taxonomists.

The problem of finding an item-set of maximal aggregated utility that satisfies a set of constraints is at the cornerstone of many search applications. Its classical definition assumes that all the information needed to verify the constraints is explicitly given. However, in real-world databases, the data available on items is often partial. Hence, adequately answering constrained search queries requires the completion of this missing information. A common approach to complete missing data is to employ Machine Learning (ML)-based inference. However, such methods are naturally error-prone. More accurate data can be obtained by asking humans to complete missing information. But, as the number of items in the repository is vast, limiting human effort is crucial. To this end, we introduce the Probabilistic Constrained Search (PCS) problem, which identifies a bounded-size item-set whose data completion is likely to be highly beneficial, as these items are expected to belong to the result set of the constrained search queries in question. We prove PCS to be hard to approximate, and consequently propose a best-effort PTIME heuristic to solve it. We demonstrate the effectiveness and efficiency of our algorithm over real-world datasets and scenarios, showing that our algorithm significantly improves the result sets of constrained search queries, in terms of both utility and constraints satisfaction probability.

Visual search has become popular in recent years, allowing users to search by an image they are taking using their mobile device or uploading from their photo library. One domain in which visual search is especially valuable is electronic commerce, where users seek for items to purchase. In this work, we present an in-depth comprehensive study of visual e-commerce search. We perform query log analysis of one of the largest e-commerce platforms' mobile search application. We compare visual and textual search by a variety of characteristics, with special focus on the retrieved results and user interaction with them. We also examine image query characteristics, refinement by attributes, and performance prediction for visual search queries. Our analysis points out a variety of differences between visual and textual e-commerce search. We discuss the implications of these differences for the design of future e-commerce search systems.

E-commerce platforms often use a predefined structured hierarchy of product categories. Apart from helping buyers sort between different product types, listing categorization is also critical for multiple downstream tasks, including the platform's main listing search. Traditionally, when creating a new listing, sellers need to assign the product they sell to a single category. However, the high diversity of product types in the platform, along with the hierarchy's low level of granularity result in tens of thousands of different possible categories that sellers need to pick from. This, in turn, creates a unique classification challenge, especially for sellers with a large number of listings. Moreover, the expected cost of making a category classification error is high, as it can impact the likelihood that their listing will get discovered by relevant buyers, and eventually sold.

Product reviews play a key role in e-commerce platforms. Studies show that many users read product reviews before purchase and trust them as much as personal recommendations. However, in many cases, the number of reviews per product is large and finding useful information becomes a challenging task. A few websites have recently added an option to post tips - short, concise, practical, and self-contained pieces of advice about products. These tips are complementary to the reviews and usually add a new non-trivial insight about the product, beyond its title, attributes, and description. Yet, most if not all major e-commerce platforms lack the notion of a tip as a first class citizen and customers typically express their advice through other means, such as reviews. In this work, we propose an extractive method for tip generation from product reviews. We focus on five popular e-commerce domains whose reviews tend to contain useful non-trivial tips that are beneficial for potential customers. We formally define the task of tip extraction in e-commerce by providing the list of tip types, tip timing (before and/or after the purchase), and connection to the surrounding context sentences. To extract the tips, we propose a supervised approach and provide a labeled dataset, annotated by human editors, over 14,000 product reviews using a dedicated tool. To demonstrate the potential of our approach, we compare different tip generation methods and evaluate them both manually and over the labeled set. Our approach demonstrates especially high performance for popular products in the Baby, Home Improvement and Sports & Outdoors domains, with precision of over 95% for the top 3 tips per product.

Search over massive sets of items is the cornerstone of many modern applications. Users express a set of properties and expect the system to retrieve qualifying items. A common difficulty, however, is that the information on whether an item satisfies the search criteria is not explicitly recorded in the repository. Instead, it may be general knowledge or "hidden" in a picture/description, leading to incomplete search results. To overcome this problem, companies build dedicated classifiers that determine which items satisfy the given criteria. However, building classifiers requires volumes of high-quality labeled training data. Since the costs of training classifiers for different subsets of properties can vastly differ, the choice of which classifiers to train has great monetary significance. The goal of our research is to devise effective algorithms to choose which classifiers one should train to address a given query load while minimizing the cost. Previous work considered a simplified model with uniform classifier costs, and queries with two properties. We remove these restrictions in our model. We prove NP-hard inapproximability bounds and devise several algorithms with approximation guarantees. Moreover, we identify a common special case for which we provide an exact algorithm. Our experiments, performed over real-life datasets, demonstrate the effectiveness and efficiency of our algorithms.

Search mechanisms over massive sets of items are the cornerstone of many modern applications, particularly in e-commerce websites. Consumers express in search queries a set of properties, and expect the system to retrieve qualifying items. A common difficulty, however, is that the information on whether or not an item satisfies the search criteria is sometimes not explicitly recorded in the repository. Instead, it may be considered as general knowledge or "hidden" in a picture/description, thereby leading to incomplete search results. To overcome these problems companies invest in building dedicated classifiers that determine whether an item satisfies the given search criteria. However, building classifiers typically incurs non-trivial costs due to the required volumes of high-quality labeled training data. In this demo, we introduce MC3, a real-time system that helps data analysts decide which classifiers to construct to minimize the costs of answering a set of search queries. MC3 is interactive and facilitates real-time analysis, by providing detailed classifiers impact information. We demonstrate the effectiveness of MC3 on real-world data and scenarios taken from a large e-commerce system, by interacting with the SIGMOD'20 audience members who act as analysts.

The problem of finding an item-set of maximal aggregated utility that satisfies a set of constraints is at the cornerstone of many e-commerce applications. Its classical definition assumes that all the information needed to verify the constraints is explicitly given. In practice, however, the data available in e-commerce databases on the items is often partial. Hence, adequately answering constrained search queries requires the completion of this missing information. A common approach to complete missing data is to employ Machine Learning (ML) algorithms. However, ML is naturally error-prone. More accurate data can be obtained by asking the items' sellers to complete missing data. But as the number of items in the repository is huge, asking sellers about all items is prohibitively expensive. CONCIERGE, our presented system, assists the e-commerce platform in identifying a bounded-size set of items whose data should be manually completed, as these items are expected to contribute the most to the constrained search queries in question. We demonstrate the effectiveness of our system on real-world data and scenarios taken from a large e-commerce system by interacting with the VLDB'20 participants who act as both analysts and the sellers.

Personalization plays a key role in electronic commerce, adjusting the products presented to users through search and recommendations according to their personality and tastes. Current personalization efforts focus on the adaptation of product selections, while the description of a given product remains the same regardless of the user who views it. In this work, we propose an approach to personalize product descriptions according to the personality of an individual user. To the best of our knowledge, we are the first to address the problem of generating personalized product descriptions. We first learn to predict a user's personality based on past activity on an e-commerce website. Then, given a user personality, we propose an extractive summarization-based algorithm that selects the sentences to be used as part of a product description in accordance with the given personality. Our evaluation shows that user personality can be effectively learned from past e-commerce activity, while personalized descriptions can lead to a higher interest in the product and increased purchase likelihood.

Many e-commerce platforms serve as an intermediary between companies/manufacturers and consumers, receiving a commission per purchase. To increase revenue, such sites tend to offer a wide variety of items. However, in many situations a smaller subset of the items should be selected and offered for sale, e.g., when opening an express branch or expanding to a new region, or when maintenance costs become prohibitive and redundant items should be disposed of. In all these cases selecting a reduced inventory which covers most consumer needs is an important goal.

Product descriptions play an important role in the e-commerce ecosystem, conveying to buyers information about a merchandise they may purchase. Yet, on leading e-commerce websites, with high volumes of new items offered for sale every day, product descriptions are often lacking or missing altogether. Moreover, many descriptions include information that holds little value and sometimes even disrupts buyers, in an attempt to draw attention and purchases. In this work, we suggest to mitigate these issues by generating short crowd-based product descriptions from user reviews . We apply an extractive approach, where review sentences are used in their original form to compose the product description. At the core of our method is a supervised approach to identify candidate review sentences suitable to be used as part of a description. Our analysis, based on data from both the Fashion and Motors domains, reveals the top reasons for review sentences being unsuitable for the product's description and these are used, in turn, as part of a deep multi-task learning architecture. We then diversify the set of candidates by removing redundancies and, at the final step, select the top candidates to be included in the description. We compare different methods for each step and also conduct an end-to-end evaluation, based on rating from professional annotators, showing the generated descriptions are of high quality.

We present WiClean, a Wikipedia plug-in that supports the identification and cleaning of certain types of errors in Wikipedia interlinks. The system mines update patterns in Wikipedia revision logs, identifies the common time frames in which they occur, and employs them to signal incomplete/inconsistent updates and suggests corrections. We demonstrate the effectiveness of WiClean in identifying actual errors in a variety of Wikipedia entity types, interactively employing the VLDB'19 audience as editors to correct the identified errors.

Writing rules to capture precisely fraudulent transactions is a challenging task where domain experts spend significant effort and time. A key observation is that much of this difficulty originates from the fact that such experts typically work as "lone rangers" or in isolated groups, or work on detecting frauds in one context in isolation from frauds that occur in another context. However, in practice there is a lot of commonality in what different experts are trying to achieve. In this paper, we present the GOLDRUSH system, which facilitates knowledge sharing via effective adaptation of fraud detection rules from one context to another. GOLDRUSH abstracts the possible semantic interpretations of each of the conditions in the rules at the source context and adapts them to the target context. Efficient algorithms are used to identify the most effective rule adaptations w.r.t a given cost-benefit metric. Our extensive set of experiments, based on real-world financial datasets, demonstrate the efficiency and effectiveness of our solution, both in terms of the accuracy of the fraud detection and the actual money saved.

Popular techniques for data cleaning use integrity constraints to identify errors in the data and to automatically resolve them, e.g. by using predefined priorities among possible updates and finding a minimal repair that will resolve violations. Such automatic solutions however cannot ensure precision of the repairs since they do not have enough evidence about the actual errors and may in fact lead to wrong results with respect to the ground truth. It has thus been suggested to use domain experts to examine the potential updates and choose which should be applied to the database.

Fraud detection rules, written by domain experts, are often employed by financial companies to enhance their machine learning-based mechanisms for accurate detection of fraudulent transactions. Accurate rule writing is a challenging task where domain experts spend significant effort and time. A key observation is that much of this difficulty originates from the fact that experts typically work as "lone rangers" or in isolated groups to define the rules, or work on detecting frauds in one context in isolation from frauds that occur in another context. However, in practice there is a lot of commonality in what different experts are trying to achieve. In this demo, we present the GOLDRUSH system, which facilitates knowledge sharing via effective adaptation of fraud detection rules from one context to another. GOLDRUSH abstracts the possible semantic interpretations of each of the conditions in the rules in one context and adapts them to the target context. Efficient algorithms are used to identify the most effective rule adaptations w.r.t a given cost-benefit metric. We showcase GOLDRUSH through a reenactment of a real-life fraud detection event. Our demonstration will engage the VLDB'18 audience, allowing them to play the role of experts collaborating in the fight against financial frauds.

We demonstrate DANCE, a system that assists domain experts in the efficient resolution of integrity constraints violation. DANCE is demonstrated on the UEFA Champions League database, employing the ICDE'17 audience for effective data cleaning.

Credit card frauds are unauthorized transactions that are made or attempted by a person or an organization that is not authorized by the card holders. In addition to machine learning-based techniques, credit card companies often employ domain experts to manually specify rules that exploit domain knowledge for improving the detection process. Over time, however, as new (fraudulent and legitimate) transaction arrive, these rules need to be updated and refined to capture the evolving (fraud and legitimate) activity patterns. The goal of the RUDOLF system that is demonstrated here is to guide and assist domain experts in this challenging task. RUDOLF automatically determines a best set of candidate adaptations to existing rules to capture all fraudulent transactions and, respectively, omit all legitimate transactions. The proposed modifications can then be further refined by domain experts based on their domain knowledge, and the process can be repeated until the experts are satisfied with the resulting rules. Our experimental results on real-life datasets demonstrate the effectiveness and efficiency of our approach. We showcase RUDOLF with two demonstration scenarios: detecting credit card frauds and network attacks. Our demonstration will engage the VLDB audience by allowing them to play the role of a security expert, a credit card fraudster, or a network attacker.

As key decisions are often made based on information contained in a database, it is important for the database to be as complete and correct as possible. For this reason, many data cleaning tools have been developed to automatically resolve inconsistencies in databases. However, data cleaning tools provide only best-effort results and usually cannot eradicate all errors that may exist in a database. Even more importantly, existing data cleaning tools do not typically address the problem of determining what information is missing from a database.

As key decisions are often made based on information contained in a database, it is important for the database to be as complete and correct as possible. For this reason, many data cleaning tools have been developed to automatically resolve inconsistencies in databases. However, data cleaning tools provide only best-effort results and usually cannot eradicate all errors that may exist in a database. Even more importantly, existing data cleaning tools do not typically address the problem of determining what information is missing from a database. To tackle these problems, we present QOCO, a novel query oriented cleaning system that leverages materialized views that are defined by user queries as a trigger for identifying the remaining incorrect/missing information. Given a user query, QOCO interacts with domain experts (which we model as oracle crowds) to identify potentially wrong or missing answers in the result of the user query, as well as determine and correct the wrong data that is the cause for the error(s). We will demonstrate QOCO over a World Cup Games database, and illustrate the interaction between QOCO and the oracles. Our demo audience will play the role of oracles, and we show how QOCO's underlying operations and optimization mechanisms can effectively prune the search space and minimize the number of questions that need to be posed to accelerate the cleaning process.

Crowd data sourcing is increasingly used to gather information from the crowd and to obtain recommendations. In this paper, we explore a novel approach that broadens crowd data sourcing by enabling users to pose general questions, to mine the crowd for potentially relevant data, and to receive concise, relevant answers that represent frequent, significant data patterns. Our approach is based on (1) a simple generic model that captures both ontological knowledge as well as the individual history or habits of crowd members from which frequent patterns are mined; (2) a query language in which users can declaratively specify their information needs and the data patterns of interest; (3) an efficient query evaluation algorithm, which enables mining semantically concise answers while minimizing the number of questions posed to the crowd; and (4) an implementation of these ideas that mines the crowd through an interactive user interface. Experimental results with both real-life crowd and synthetic data demonstrate the feasibility and effectiveness of the approach.

We present OASSIS (for Ontology ASSISted crowd mining), a prototype system which allows users to declaratively specify their information needs, and mines the crowd for answers. The answers that the system computes are concise and relevant , and represent frequent, significant data patterns. The system is based on (1) a generic model that captures both ontological knowledge, as well as the individual knowledge of crowd members from which frequent patterns are mined; (2) a query language in which users can specify their information needs and types of data patterns they seek; and (3) an efficient query evaluation algorithm, for mining semantically concise answers while minimizing the number of questions posed to the crowd. We will demonstrate OASSIS using a couple of real-life scenarios, showing how users can formulate and execute queries through the OASSIS UI and how the relevant data is mined from the crowd.

Recent research has shown that crowd sourcing can be used effectively to solve problems that are difficult for computers, e.g., optical character recognition and identification of the structural configuration of natural proteins [1]. In this demo we propose to use the power of the crowd to address yet another difficult problem that frequently occurs in a daily life-planning a sequence of actions, when the goal is hard to formalize. For example, planning the sequence of places/attractions to visit in the course of a vacation, where the goal is to enjoy the resulting vacation the most, or planning the sequence of courses to take in an academic schedule planning, where the goal is to obtain solid knowledge of a given subject domain. Such goals may be easily understandable by humans, but hard or even impossible to formalize for a computer. We present a novel algorithm for efficiently harnessing the crowd to assist in solving such planning problems. The algorithm builds the desired plans incrementally, optimally choosing at each step the `best' questions so that the overall number of questions that need to be asked is minimized. We demonstrate the effectiveness of our solution in CrowdPlanr, a system for vacation travel planning. Given a destination, dates, preferred activities and other constraints CrowdPlanr employs the crowd to build a vacation plan (sequence of places to visit) that is expected to maximize the “enjoyment” of the vacation.

Recent research has shown that crowd sourcing can be used effectively to solve problems that are difficult for computers, e.g., optical character recognition and identification of the structural configuration of natural proteins. In this paper we propose to use the power of the crowd to address yet another difficult problem that frequently occurs in a daily life - answering planning queries whose output is a sequence of objects/actions, when the goal, i.e, the notion of "best output", is hard to formalize. For example, planning the sequence of places/attractions to visit in the course of a vacation, where the goal is to enjoy the resulting vacation the most, or planning the sequence of courses to take in an academic schedule planning, where the goal is to obtain solid knowledge of a given subject domain. Such goals may be easily understandable by humans, but hard or even impossible to formalize for a computer. We present a novel algorithm for efficiently harnessing the crowd to assist in answering such planning queries. The algorithm builds the desired plans incrementally, choosing at each step the 'best' questions so that the overall number of questions that need to be asked is minimized. We prove the algorithm to be optimal within its class and demonstrate experimentally its effectiveness and efficiency.

Crowd-based data sourcing is a new and powerful data procurement paradigm that engages Web users to collectively contribute information. In this work, we target the problem of gathering data from the crowd in an economical and principled fashion. We present Ask It!, a system that allows interactive data sourcing applications to effectively determine which questions should be directed to which users for reducing the uncertainty about the collected data. Ask It! uses a set of novel algorithms for minimizing the number of probing (questions) required from the different users. We demonstrate the challenge and our solution in the context of a multiple-choice question game played by the ICDE'12 attendees, targeted to gather information on the conference's publications, authors and colleagues.

This paper considers a popular class of recommender systems that are based on Collaborative Filtering (CF) and proposes a novel technique for diversifying the recommendations that they give to users. Items are clustered based on a unique notion of priority-medoids that provides a natural balance between the need to present highly ranked items vs. highly diverse ones. Our solution estimates items diversity by comparing the rankings that different users gave to the items, thereby enabling diversification even in common scenarios where no semantic information on the items is available. It also provides a natural zoom-in mechanism to focus on items (clusters) of interest and recommending diversified similar items. We present DiRec a plug-in that implements the above concepts and allows CF Recommender systems to diversify their recommendations. We illustrate the operation of DiRec in the context of a movie recommendation system and present a thorough experimental study that demonstrates the effectiveness of our recommendation diversification technique and its superiority over previous solutions.

In this demo we present DiRec, a plug-in that allows Collaborative Filtering (CF) Recommender systems to diversify the recommendations that they present to users. DiRec estimates items diversity by comparing the rankings that different users gave to the items, thereby enabling diversification even in common scenarios where no semantic information on the items is available. Items are clustered based on a novel notion of priority-medoids that provides a natural balance between the need to present highly ranked items vs. highly diverse ones. We demonstrate the operation of DiRec in the context of a movie recommendation system. We show the advantage of recommendation diversification and its feasibility even in the absence of semantic information.