Recommender systems used in current online social platforms make recommendations by only considering how relevant an item is to a specific user but they ignore the fact that, thanks to mechanisms like sharing or re-posting across the underlying social network, an item recommended to a user i propagates through the network and can reach another user j without needing to be explicitly recommended to j too. Overlooking this fact may lead to an inefficient use of the limited recommendation slots. These slots can instead be exploited more profitably by avoiding unnecessary duplicates and recommending other equally relevant items.

Extraction of patterns out of streaming data that are generated from geographically dispersed devices is a major challenge in data mining. The sequential, distributed fashion in which data become available to the decision maker, together with the fact that the decision maker needs to rely only on recently received data due to storage and communication constraints, render the objective of keeping track of data evolution a nontrivial one. We consider a set of distributed nodes that communicate directly with a central location. We address the problem of clustering distributed streaming data through a two-level clustering approach. We adopt belief propagation techniques to perform stream clustering at both levels. At the node level, a batch of data arrives at each time slot, and the goal is to maintain a set of salient data (local exemplars) at each time slot, which best represents the data received up to that slot. At each epoch, the local exemplars from distributed nodes are sent to the central location, which in turn performs a second-level clustering on them to derive a data synopsis global for the whole system. The local exemplars that emerge from the second level clustering procedure are fed back to the nodes with appropriately modified weights which reflect their importance in global clustering. As demonstrated by our experiments, the two-level belief propagation-based clustering approach together with the feedback is ideal for handling data from different nodes, as it has the same performance in terms of clustering quality with the case where the clustering is performed on the raw data sent from nodes to the central location.

The increased availability of data created as part of the software development process allows us to apply novel analysis techniques on the data and use the results to guide the process's optimization. In this paper we describe various data sources an

We study the problem of maximum lifetime in wireless sensor networks that are entitled with the task of estimating an unknown parameter or process. Sensors take measurements and transfer them in multi-hop fashion to a fusion center (FC) for Maximum Likelihood (ML) estimation. To engineer the network for lifetime maximization while adhering to estimation error specifications, the number of measurements by each sensor per unit of time (namely, sensor measurement rate) and the routes to the FC are controlled. Sensor spatial correlation, measurement accuracies, link qualities and energy reserves affect sensor measurement rates and the routes to the FC, and, in turn, measurement rates and sensor characteristics impact estimation error. We show that the problem can be decomposed into separate optimization problems where each sensor autonomously takes its measurement rate and routing decisions, and we propose an iterative primal-dual algorithm with low-overhead signaling for solving it. Our work optimally captures the fundamental tradeoff between network lifetime and estimation quality and yields a solution based on distributed sensor coordination.

Clustering, as an unsupervised learning process is a challenging problem, especially in cases of high-dimensional datasets. Clustering result quality can benefit from user constraints and objective validity assessment. In this article, we propose a semisupervised framework for learning the weighted Euclidean subspace, where the best clustering can be achieved. Our approach capitalizes on: (i) user constraints; and (ii) the quality of intermediate clustering results in terms of their structural properties. The proposed framework uses the clustering algorithm and the validity measure as its parameters. We develop and discuss algorithms for learning and tuning the weights of contributing dimensions and defining the “best” clustering obtained by satisfying user constraints. Experimental results on benchmark datasets demonstrate the superiority of the proposed approach in terms of improved clustering accuracy.

Sensor networks are being deployed for tracking events of interest in many environmental or monitoring applications. Because of their distributed nature of operation, a challenging issue is how to accurately identify the aggregate state of the phenomenon that is being observed. This work presents an online mechanism for efficiently determining the overall network status, employing distributed operations that minimize the communication costs. Experiments on real data, suggest that the proposed metholology can be a viable solution for real world systems.

In this paper, we propose a semi-supervised framework for learning a weighted Euclidean subspace, where the best clustering can be achieved. Our approach capitalizes on user-constraints and the quality of intermediate clustering results in terms of its structural properties. It uses the clustering algorithm and the validity measure as parameters.

With the unstoppable growth of the world wide Web, the great success of Web search engines, such as Google and AltaVista, users now turn to the Web whenever looking for information. However, many users are neophytes when it comes to computer science, yet they are often specialists of a certain domain. These users would like to add more semantics to guide their search through world wide Web material, whereas currently most search features are based on raw lexical content. We show how the use of the incoming links of a page can be used efficiently to classify a page in a concise manner. This enhances the browsing and querying of Web pages. We focus on the tools needed in order to manage the links and their semantics. We further process these links using a hierarchy of concepts, akin to an ontology, and a thesaurus. This work is demonstrated by an prototype system, called THESUS, that organizes thematic Web documents into semantic clusters. Our contributions are the following: 1) a model and language to exploit link semantics information, 2) the THESUS prototype system, 3) its innovative aspects and algorithms, more specifically, the novel similarity measure between Web documents applied to different clustering schemes (DB-Scan and COBWEB), and 4) a thorough experimental evaluation proving the value of our approach.

Clustering is a mostly unsupervised procedure and the majority of clustering algorithms depend on certain assumptions in order to define the subgroups present in a data set. As a consequence, in most applications the resulting clustering scheme requires some sort of evaluation regarding its validity. In this paper we present a clustering validity procedure, which evaluates the results of clustering algorithms on data sets. We define a validity index, S Dbw, based on well-defined clustering criteria enabling the selection of optimal input parameter values for a clustering algorithm that result in the best partitioning of a data set. We evaluate the reliability of our index both theoretically and experimentally, considering three representative clustering algorithms run on synthetic and real data sets. We also carried out an evaluation study to compare S Dbw performance with other known validity indices. Our approach performed favorably in all cases, even those in which other indices failed to indicate the correct partitions in a data set.

Clustering aims at discovering groups and identifying interesting distributions and patterns in data sets. Researchers have extensively studied clustering since it arises in many application domains in engineering and social sciences. In the last years the availability of huge transactional and experimental data sets and the arising requirements for data mining created needs for clustering algorithms that scale and can be applied in diverse domains. The paper surveys clustering methods and approaches available in the literature in a comparative way. It also presents the basic concepts, principles and assumptions upon which the clustering algorithms are based. Another important issue is the validity of the clustering schemes resulting from applying algorithms. This is also related to the inherent features of the data set under concern. We review and compare clustering validity measures available in the literature. Furthermore, we illustrate the issues that are under-addressed by the recent algorithms and we address new research directions.

The KDD process aims at searching for interesting instances of patterns in data sets. It is widely accepted that the patterns must be comprehensible. One of the aspects that are under-addressed in the KDD process is the handling of uncertainty in the process of clustering, classification and association rules extraction. In this paper we present a classification framework for relational databases so as to support uncertainty in terms of natural language queries and assessments. More specifically, we present a classification scheme of non-categorical attributes into lexically defined categories based on fuzzy logic and provides decision support facilities based on related information measures.