The ACM Recommender Systems Challenge 20171 focused on the problem of job recommendations: given a new job advertisement, the goal was to identify those users who are both (a) interested in getting notified about the job advertisement, and (b) appropriate candidates for the given job. Participating teams had to balance between user interests and requirements for the given job as well as dealing with the cold-start situation. For the first time in the history of the conference, the RecSys challenge offered an online evaluation: teams first had to compete as part of a traditional offline evaluation and the top 25 teams were then invited to evaluate their algorithms in an online setting, where they could submit recommendations to real users. Overall, 262 teams registered for the challenge, 103 teams actively participated and submitted together more than 6100 solutions as part of the offline evaluation. Finally, 18 teams participated and rolled out recommendations to more than 900,000 users on XING2.

The 2016 ACM Recommender Systems Challenge focused on the problem of job recommendations. Given a large dataset from XING that consisted of anonymized user profiles, job postings, and interactions between them, the participating teams had to predict postings that a user will interact with. The challenge ran for four months with 366 registered teams. 119 of those teams actively participated and submitted together 4,232 solutions yielding in an impressive neck-and-neck race that was decided within the last days of the challenge.

Business-oriented social networks like LinkedIn or XING support people in discovering career opportunities. In this talk, we will focus on the problem of recommending job offers to Millions of XING users. We will discuss challenges of building a job recommendation system that has to satisfy the demands of both job seekers who have certain wishes concerning their next career step and recruiters who aim to hire the most appropriate candidate for a job. Based on insights gained from a large-scale analysis of usage data and profile data such as curriculum vitae, we will study features of the recommendation algorithms that aim to solve the problem.

Conducting analytics over data generated by Social Web portals such as Twitter is challenging, due to the volume, variety and velocity of the data. Commonly, adhoc pipelines are used that solve a particular use case. In this paper, we generalize across a range of typical Twitter-data use cases and determine a set of common characteristics. Based on this investigation, we present our Twitter Analytical Platform (TAP), a generic platform for conducting analytical tasks with Twitter data. The platform provides a domain-specific Twitter Analysis Language (TAL) as the interface to its functionality stack. TAL includes a set of analysis tools ranging from data collection and semantic enrichment, to machine learning. With these tools, it becomes possible to create and customize analytical workflows in TAL and build applications that make use of the analytics results. We showcase the applicability of our platform by building Twinder-a search engine for Twitter streams.

With more than 340~million messages that are posted on Twitter every day, the amount of duplicate content as well as the demand for appropriate duplicate detection mechanisms is increasing tremendously. Yet there exists little research that aims at detecting near-duplicate content on microblogging platforms. We investigate the problem of near-duplicate detection on Twitter and introduce a framework that analyzes the tweets by comparing (i) syntactical characteristics, (ii) semantic similarity, and (iii) contextual information. Our framework provides different duplicate detection strategies that, among others, make use of external Web resources which are referenced from microposts. Machine learning is exploited in order to learn patterns that help identifying duplicate content. We put our duplicate detection framework into practice by integrating it into Twinder, a search engine for Twitter streams. An in-depth analysis shows that it allows Twinder to diversify search results and improve the quality of Twitter search. We conduct extensive experiments in which we (1) evaluate the quality of different strategies for detecting duplicates, (2) analyze the impact of various features on duplicate detection, (3) investigate the quality of strategies that classify to what exact level two microposts can be considered as duplicates and (4) optimize the process of identifying duplicate content on Twitter. Our results prove that semantic features which are extracted by our framework can boost the performance of detecting duplicates.

In this work we present an in-depth analysis of the user behaviors on different Social Sharing systems. We consider three popular platforms, Flickr, Delicious and StumbleUpon, and, by combining techniques from social network analysis with techniques from semantic analysis, we characterize the tagging behavior as well as the tendency to create friendship relationships of the users of these platforms. The aim of our investigation is to see if (and how) the features and goals of a given Social Sharing system reflect on the behavior of its users and, moreover, if there exists a correlation between the social and tagging behavior of the users. We report our findings in terms of the characteristics of user profiles according to three different dimensions: (i) intensity of user activities, (ii) tag-based characteristics of user profiles, and (iii) semantic characteristics of user profiles.

In this paper, we study user modeling on Twitter and investigate the interplay between personal interests and public trends. To generate semantically meaningful user profiles, we present a framework that allows us to enrich the semantics of individual Twitter messages and features user modeling as well as trend modeling strategies. These profiles can be re-used in other applications for (trend-aware) personalization. Given a large Twitter dataset, we analyze the characteristics of user and trend profiles and evaluate the quality of the profiles in the context of a personalized news recommendation system. We show that personal interests are more important for the recommendation process than public trends and that by combining both types of profiles we can further improve recommendation quality.

Over the last decade blogs became an important part of the Web, where people can announce anything that is on their mind. Due to their high popularity blogs have great potential to mine public opinions regarding products. Such knowledge is very valuable as it could be used to adjust marketing campaigns or advertisement of products accordingly. In this paper we investigate how the blogosphere can be used to predict the success of products in the domain of music and movies. We analyze and characterize the blogging behavior in both domains particularly around product releases, propose different methods for extracting characteristic features from the blogosphere, and show that our predictions correspond to the real world measures Sales Rank and box office revenue respectively.

Folksonomy systems have shown to contribute to the quality of Web search ranking strategies. In this paper, we analyze and compare different graph-based ranking algorithms, namely FolkRank, SocialPageRank, and SocialSimRank. We enhance these algorithms by exploiting the context of tag assignmets, and evaluate the results on the GroupMe! dataset. In GroupMe!, users can organize and maintain arbitrary Web resources in self-defined groups. When users annotate resources in GroupMe!, this can be interpreted in context of a certain group. The grouping activity delivers valuable semantic information about resources and their context. We show how to use this information to improve the detection of relevant search results, and compare different strategies for ranking result lists in folksonomy systems.

Folksonomy systems enable users to participate in the Web content creation process by annotating (tagging) resources with freely chosen keywords. Still, it is an open issue how to exploit this user-created content, and how to process and use these emergent semantics effectively. We investigate how the context of Web resources can be utilized to improve recommended strategies in social tagging systems. We focus on the GroupMe! system, which enables users to create groups in order to bundle Web resources. GroupMe! groups form valuable context information for the resources contained in such groups. In this paper we exploit graph-based tag recommendation strategies, evaluate them in the GroupMe! dataset, and benchmark our results against other approaches. In our evaluations we show that graph-based strategies outperform other approaches, and show the immanent benefit of graph-based recommendation strategies which exploit the group context for recommending tags to untagged resources.