An important aspect of music perception in humans is the ability to segment streams of musical events into structural units such as motifs and phrases.A promising approach to the computational modeling of music segmentation employs the statistical and information-theoretic properties of musical data, based on the hypothesis that these properties can (at least partly) account for music segmentation in humans. Prior work has shown that in particular the information content of music events, as estimated from a generative probabilistic model of those events, is a good indicator for segment boundaries.In this paper we demonstrate that, remarkably, a substantial increase in segmentation accuracy can be obtained by not using information content estimates directly, but rather in a bootstrapping fashion. More specifically, we use information content estimates computed from a generative model of the data as a target for a feed-forward neural network that is trained to estimate the information content directly from the data. We hypothesize that the improved segmentation accuracy of this bootstrapping approach may be evidence that the generative model provides noisy estimates of the information content, which are smoothed by the feed-forward neural network, yielding more accurate information content estimates.

Artistic creation is often based on the concept of blending. Linguistic creativity is no exception, as demonstrated for instance by the importance of metaphors in poetry. Blending can also be used to evoke a secondary concept while playing with an already given piece of language, either with the intention of making the secondary concept well perceivable to the reader, or instead, to subtly evoke something additional. Current language technology can do a lot in this connection, and automated language creativity can be useful in cases where input or target are to change continuously, making human production not feasible. In this work we present a system that takes existing well-known expressions and innovates them by bringing in a novel concept coming from evolving news. The technology is composed of several steps concerned with the selection of the sortable concepts and the production of novel expressions, largely relying on state of the art corpus-based methods. Proposed applications include: i) producing catchy news headlines by "parasitically" exploiting well known successful expressions and adapting them to the news at hand; ii) generating adaptive slogans that allude to news of the day and give life to the concept evoked by the slogan; iii) providing artists with an application for boosting their creativity.

We present a computational framework for chord invention based on a cognitive-theoretic perspective on conceptual blending. The framework builds on algebraic specifications, and solves two musicological problems. It automatically finds transitions between chord progressions of different keys or idioms, and it substitutes chords in a chord progression by other chords of a similar function, as a means to create novel variations. The approach is demonstrated with several examples where jazz cadences are invented by blending chords in cadences from earlier idioms, and where novel chord progressions are generated by inventing transition chords.

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.

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.

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.

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.

In this paper we perform a large-scale homophily analysis on Twitter using a hierarchical representation of users' interests which we call a Twixonomy. In order to build a population, community, or single-user Twixonomy we first associate "topical" friends in users' friendship lists (i.e. friends representing an interest rather than a social relation between peers) with Wikipedia categories. A word-sense disambiguation algorithm is used to select the appropriate wikipage for each topical friend. Starting from the set of wikipages representing "primitive" interests, we extract all paths connecting these pages with topmost Wikipedia category nodes, and we then prune the resulting graph G efficiently so as to induce a direct acyclic graph. This graph is the Twixonomy. Then, to analyze homophily, we compare different methods to detect communities in a peer friends Twitter network, and then for each community we compute the degree of homophily on the basis of a measure of pairwise semantic similarity.We show that the Twixonomy provides a means for describing users' interests in a compact and readable way and allows for a fine-grained homophily analysis. Furthermore, we show that mid-low level categories in the Twixonomy represent the best balance between informativeness and compactness of the representation.

We propose a deep learning method for eventdriven stock market prediction. First, events are extracted from news text, and represented as dense vectors, trained using a novel neural tensor network. Second, a deep convolutional neural network is used to model both short-term and long-term influences of events on stock price movements. Experimental results show that our model can achieve nearly 6% improvements on S&P 500 index prediction and individual stock prediction, respectively, compared to state-of-the-art baseline methods. In Figure 1: Example news influence of Google Inc. addition, market simulation results show that our system is more capable of making profits than previously reported systems trained on S&P 500 stock of events can be better captured [Ding et al., 2014].

The Nystrom method provides an efficient sampling approach for large scale clustering problems, by generating a low-rank matrix approximation. However, existing sampling methods are limited by accuracy and computing time. This paper proposes an improved Nystrom-based clustering algorithm with a new sampling procedure, Minimum Sum of Squared Similarities (MSSS). Experiments on synthetic and real data sets show that the proposed sampling performs with higher accuracy than existing algorithms, applied to Nystrom-based spectral clustering problems. Furthermore, we provide a theoretical analysis that allows us to define the upper bound of the Frobenius norm error of the MSSS.