In this paper we present a real-world application (the first of its kind) of machine listening in the context of a live concert in a world-famous concert hall - the Concertgebouw in Amsterdam. A real-time music tracking algorithm listens to the Royal Concertgebouw Orchestra performing Richard Strauss' Alpensinfonie and follows the progress in the sheet music, i.e., continuously tracks the most likely position of the live music in the printed score. This information, in turn, is used to enrich the concert experience for members of the audience by streaming synchronised visual content (the sheet music, explanatory text and videos) onto tablet computers in the concert hall. The main focus of this paper is on the challenges involved in tracking live orchestral music, i.e., how to deal with heavily polyphonic music, how to prepare the data needed, and how to achieve the necessary robustness and precision.

Recommendation systems play an important role in E-Commerce. However, their potential usefulness in real world applications is greatly limited by the availability of historical rating records from the customers. This paper presents a novel method to tackle the problem of data sparseness in user ratings with rich and timely domain information from social media. We first extract multiple side information for products from their relevant social media contents. Next, we convert the information into weighted topic-item ratings and inject them into an extended latent factor based recommendation model in an optimized approach. Our evaluation on two real world datasets demonstrates the superiority of our method over state-of-the-art methods.

Many e-commerce systems allow users to express their opinions towards products through user reviews systems. The user generated reviews not only help other users to gain a more insightful view of the products, but also help online businesses to make targeted improvements on the products or services. Besides, they compose the key component of various personalized recommender systems. However, the existence of spam user accounts in the review systems introduce unfavourable disturbances into personalized recommendation by promoting or degrading targeted items intentionally through fraudulent reviews. Previous shilling attack detection algorithms usually deal with a specific kind of attacking strategy, and are exhausted to handle with the continuously emerging new cheating methods. In this work, we propose to conduct shilling attack detection for more informed recommendation by fraudulent action propagation on the reviews themselves, without caring about the specific underlying cheating strategy, which allows us a unified and flexible framework to detect the spam users.

Image location prediction is to estimate the geolocation where an image is taken. Social image contains heterogeneous contents, which makes image location prediction nontrivial. Moreover, it is observed that image content patterns and location preferences correlate hierarchically. Traditional image location prediction methods mainly adopt a single-level architecture, which is not directly adaptable to the hierarchical correlation. In this paper, we propose a geographically hierarchical bi-modal deep belief network model (GH-BDBN), which is a compositional learning architecture that integrates multi-modal deep learning model with non-parametric hierarchical prior model. GH-BDBN learns a joint representation capturing the correlations among different types of image content using a bi-modal DBN, with a geographically hierarchical prior over the joint representation to model the hierarchical correlation between image content and location. Experimental results demonstrate the superiority of our model for image location prediction.

Community detection is of crucial importance in understanding structures of complex networks. In many real-world networks, communities naturally overlap since a node usually has multiple community memberships. One popular technique to cope with overlapping community detection is Matrix Factorization (MF). However, existing MF-based models have ignored the fact that besides neighbors, "local non-neighbors" (e.g., my friend's friend but not my direct friend) are helpful when discovering communities. In this paper, we propose a Locality-based Non-negative Matrix Factorization (LNMF) model to refine a preference-based model by incorporating locality into learning objective. We define a subgraph called "k-degree local network" to set a boundary between local non-neighbors and other non-neighbors. By discriminately treating these two class of non-neighbors, our model is able to capture the process of community formation. We propose a fast sampling strategy within the stochastic gradient descent based learning algorithm. We compare our LNMF model with several baseline methods on various real-world networks, including large ones with ground-truth communities. Results show that our model outperforms state-of-the-art approaches.

In this work, we investigate the bidirectional mutual interactions (BMI) between users' activities and user-user relationships on social networking sites. We analyze and study the fundamental mechanism that drives the characteristics and dynamics of BMI is the underlying social influence. We make an attempt at a unified probabilistic approach, called joint activity and relation (JAR), for modeling and predicting users' activities and user-user relationships simultaneously in a single coherent framework. Instead of incorporating social influence in an ad hoc manner, we show that social influence can be captured quantitatively. Based on JAR, we learn social influence between users and users' personal preferences for both user activity prediction and user-user relation discovery through statistical inference. To address the challenges of the introduced multiple layers of hidden variables in JAR, we propose a new learning algorithm based on expectation maximization (EM) and we further propose a powerful and efficient generalization of the EM based algorithm for model fitting.We show that JAR exploits mutual interactions and benefits, by taking advantage of the learned social influence and users' personal preferences, for enhanced user activity prediction and user-user relation discovery. We further experiment with real world dataset to verify the claimed advantages achieving substantial performance gains.

The vote mechanism is widely utilized to rank answers in community-based question answering sites. In generating a vote, a user's attention is influenced by the answer position and appearance, in addition to real answer quality. Previously, these biases are ignored. As a result, the top answers obtained from this mechanism are not reliable, if the number of votes for the active question is not sufficient. In this paper, we solve this problem by analyzing two kinds of biases; position bias and appearance bias. We identify the existence of these biases and propose a joint click model for dealing with both of them. Our experiments in real data demonstrate how the ranking performance of the proposed model outperforms traditional methods with biases ignored by 15.1% in precision@1, and 11.7% in the mean reciprocal rank. A case study on a manually labeled dataset futher supports the effectiveness of the proposed model.

User interest inference from social networks is a fundamental problem to many applications. It usually exhibits dual-heterogeneities: a user's interests are complementarily and comprehensively reflected by multiple social networks; interests are inter-correlated in a nonuniform way rather than independent to each other. Although great success has been achieved by previous approaches, few of them consider these dual-heterogeneities simultaneously. In this work, we propose a structure-constrained multi-source multi-task learning scheme to co-regularize the source consistency and the tree-guided task relatedness. Meanwhile, it is able to jointly learn the task-sharing and task-specific features. Comprehensive experiments on a real-world dataset validated our scheme. In addition, we have released our dataset to facilitate the research communities.

Semantic relatedness (SR) measures form the algorithmic foundation of intelligent technologies in domains ranging from artificial intelligence to human-computer interaction. Although SR has been researched for decades, this work has focused on developing general SR measures rooted in graph and text mining algorithms that perform reasonably well for many different types of concepts. This paper introduces domain-specific SR, which augments general SR by identifying, capturing, and synthesizing domain-specific relationships between concepts. Using the domain of geography as a case study, we show that domain-specific SR — and even geography-specific signals alone (e.g. distance, containment) without sophisticated graph or text mining algorithms — significantly outperform the SR state-of-the-art for geographic concepts. In addition to substantially improving SR measures for geospatial technologies, an area that is rapidly increasing in importance, this work also unlocks an important new direction for SR research: SR measures that incorporate domain-specific customizations to increase accuracy.

There are lots of texts appearing in the web every day. This fact enables the amount of texts in the web to explode. Therefore, how to deal with large-scale text collection becomes more and more important. Clustering is a generally acceptable solution for text organization. Via its unsupervised characteristic, users can easily dig the useful information that they desired. However, traditional clustering algorithms can only deal with small-scale text collection. When it enlarges, they lose their performances. The main reason attributes to the high-dimensional vectors generated from texts. Therefore, to cluster texts in large amount, this paper proposes a novel clustering algorithm, where only the features that can represent cluster are preserved in cluster’s vector. In this algorithm, clustering process is separated into two parts. In one part, feature’s weight is fine-tuned to make cluster partition meet an optimization function. In the other part, features are reordered and only the useful features that can represent cluster are kept in cluster’s vector. Experimental results demonstrate that our algorithm obtains high performance on both small-scale and large-scale text collections.