Search

Motivated by a network fault detection problem, we study how recall can be boosted in a decision tree classifier, without sacrificing too much precision. This problem is relevant and novel in the context of transfer learning (TL), in which few target domain training samples are available. We define a geometric optimization problem for boosting the recall of a decision tree classifier, and show it is NP-hard. To solve it efficiently, we propose several near-linear time heuristics, and experimentally validate these heuristics in the context of TL. Our evaluation includes 7 public datasets, as well as 6 network fault datasets, and we compare our heuristics with several existing TL algorithms, as well as exact mixed integer linear programming (MILP) solutions to our optimization problem. We find that our heuristics boost recall in a manner similar to optimal MILP solutions, yet require several orders of magnitude less compute time. In many cases the

Time division multiple access (TDMA) is the medium access control strategy of choice for multihop networks with deterministic delay guarantee requirements. As such, many Internet of Things applications use protocols based on time division multiple access. Optimal slot assignment in such networks is NP-hard when there are strict deadline requirements and is generally done using heuristics that give suboptimal transmission schedules in linear time. However, existing heuristics make a scheduling decision at each time slot based on the same criterion without considering its effect on subsequent network states or scheduling actions. Here, we first identify a set of node features that capture the information necessary for network state representation to aid building schedules using Reinforcement Learning (RL). We then propose three different centralized approaches to RL-based TDMA scheduling that vary in training and network representation methods. Using RL allows applying diverse criteria at different time slots while considering the effect of a scheduling action on meeting the scheduling objective for the entire TDMA frame, resulting in better schedules. We compare the three proposed schemes in terms of how well they meet the scheduling objectives and their applicability to networks with memory and time constraints. One of the schemes proposed is RLSchedule, which is particularly suited to constrained networks. Simulation results for a variety of network scenarios show that RLSchedule reduces the percentage of packets missing deadlines by up to 60% compared to the best available baseline heuristic.

Nowadays, the large-scale human activity traces on social media platforms such as Twitter provide new opportunities for various research areas such as mining user interests, understanding user behaviors, or conducting social science studies in a large scale. However, social media platforms contain not only individual accounts but also other accounts that are associated with non-individuals such as organizations or brands. Therefore, distinguishing individuals out of all accounts is crucial when we conduct research such as understanding human behavior based on data retrieved from those platforms. In this paper, we propose a language-independent approach for distinguishing individuals from non-individuals with the focus on leveraging their profile images, which has not been explored in previous studies. Extensive experiments on two datasets show that our proposed approach can provide competitive performance with state-of-the-art language-dependent methods, and outperforms alternative language-independent ones.

The adoption of fast-paced practices for developing virtual network functions (VNFs) allows for continuous software delivery and creates a market advantage for network operators. This adoption, however, is problematic for testing engineers that need to assure, in shorter development cycles, certain quality of highly-configurable product releases running on heterogeneous clouds. Machine learning (ML) can accelerate testing workflows by detecting performance issues in new software builds. However, the overhead of maintaining several models for all combinations of build types, network configurations, and other stack parameters, can quickly become prohibitive and make the application of ML infeasible. We propose Env2Vec, a deep learning architecture that combines contextual features with historical resource usage, and characterizes the various stack parameters that influence the test execution within an embedding space, which allows it to generalize model predictions to previously unseen environments. We integrate a single ML model in the testing workflow to automatically debug errors and pinpoint performance bottlenecks. Results obtained with real testing data show an accuracy between 86.2%-100%, while reducing the false alarm rate by 20.9%-38.1% when reporting performance issues compared to state-of-the-art approaches.

Social media users readily share their preferences, life events, sentiment and opinions, and implicitly signal their thoughts, feelings, and psychological behavior. This makes social media a viable source of information to accurately and effectively mine users' interests with the hopes of enabling more effective user engagement, better quality delivery of appropriate services and higher user satisfaction. In this tutorial, we cover five important aspects related to the effective mining of user interests: (1) the foundations of social user interest modeling, such as information sources, various types of representation models and temporal features, (2) techniques that have been adopted or proposed for mining user interests, (3) different evaluation methodologies and benchmark datasets, (4) different applications that have been taking advantage of user interest mining from social media platforms, and (5) existing challenges, open research questions and exciting opportunities for further work.

User modeling for inferring user interests from Online Social Networks (OSNs) such as Twitter has received great attention in the user modeling community with the growing popularity of OSNs. The focus of previous works has been on analyzing user-generated content such as tweets to infer user interests. Therefore, these previous studies were limited to active users who have been actively generating content. On the other hand, with the percentage of passive use of OSNs on the rise, some researchers investigated different types of information about followees (i.e., people that a user is following) such as tweets, usernames, and biographies to infer user interests for passive users who use OSNs for consuming information from followees but who do not produce any content. Although different types of information about followees have been exploited, list memberships (a topical list which other Twitter users can freely add a user into) of followees have not yet been investigated extensively for inferring user interests. In this paper, we investigate list memberships of followees, to infer interest profiles for passive users. To this end, we propose user modeling strategies with two different weighting schemes as well as a refined interest propagation strategy based on previous work. In addition, we investigate whether the information from biographies and list memberships of followees can complement each other, and thus improve the quality of inferred interest profiles for passive users. Results show that leveraging list memberships of followees is useful for inferring user interests when the number of followees is relatively small compared to using biographies of followees. In addition, we found that combining the two different types of information (list memberships and biographies) of followees can improve the quality of user interest profiles significantly compared to a state-of-art method in the context of link recommendations on Twitter.

The main contribution of this work is the comparison of three user modeling strategies based on job titles, educational fields and skills in LinkedIn profiles, for personalized MOOC recommendations in a cold start situation. Results show that the skill-based user modeling strategy performs best, followed by the job- and edu-based strategies.

User modeling for individual users on the Social Web plays an important role and is a fundamental step for personalization as well as recommendations. Recent studies have proposed different user modeling strategies considering various dimensions such as temporal dynamics and semantics of user interests. Although previous work proposed different user modeling strategies considering the temporal dynamics of user interests, there is a lack of comparative studies on those methods and therefore the comparative performance over each other is unknown. In terms of semantics of user interests, background knowledge from DBpedia has been explored to enrich user interest profiles so as to reveal more information about users. However, it is still unclear to what extent different types of information from DBpedia contribute to the enrichment of user interest profiles. In this paper, we propose user modeling strategies which use Concept Frequency - Inverse Document Frequency (CF-IDF) as a weighting scheme and incorporate either or both of the dynamics and semantics of user interests. To this end, we first provide a comparative study on different user modeling strategies considering the dynamics of user interests in previous literature to present their comparative performance. In addition, we investigate different types of information (i.e., categories, classes and connected entities via various properties) for entities from DBpedia and the combination of them for extending user interest profiles. Finally, we build our user modeling strategies incorporating either or both of the best performing methods in each dimension. Results show that our strategies outperform two baseline strategies significantly in the context of link recommendations on Twitter.

The Linked Open Data (LOD) initiative has been quite successful in terms of publishing and interlinking data on the Web. On top of the huge amount of interconnected data, measuring relatedness between resources and identifying their relatedness could be used for various applications such as LOD-enabled recommender systems. In this paper, we propose various distance measures, on top of the basic concept of Linked Data Semantic Distance (LDSD), for calculating Linked Data semantic distance between resources that can be used in a LOD-enabled recommender system. We evaluated the distance measures in the context of a recommender system that provides the top-N recommendations with baseline methods such as LDSD. Results show that the performance is significantly improved by our proposed distance measures incorporating normalizations that use both of the resources and global appearances of paths in a graph.

User modeling of individual users on the Social Web platforms such as Twitter plays a significant role in providing personalized recommendations and filtering interesting information from social streams. Recently, researchers proposed the use of concepts (e.g., DBpedia entities) for representing user interests instead of word-based approaches, since Knowledge Bases such as DBpedia provide cross-domain background knowledge about concepts, and thus can be used for extending user interest profiles. Even so, not all concepts can be covered by a Knowledge Base, especially in the case of microblogging platforms such as Twitter where new concepts/topics emerge everyday. In this short paper, instead of using concepts alone, we propose using synsets from WordNet and concepts from DBpedia for representing user interests. We evaluate our proposed user modeling strategies by comparing them with other bag-of-concepts approaches. The results show that using synsets and concepts together for representing user interests improves the quality of user modeling significantly in the context of link recommendations on Twitter.