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.

Hashing has been widely used for approximate nearest neighbor (ANN) search in big data applications because of its low storage cost and fast retrieval speed. The goal of hashing is to map the data points from the original space into a binary-code space where the similarity (neighborhood structure) in the original space is preserved. By directly exploiting the similarity to guide the hashing code learning procedure, graph hashing has attracted much attention. However, most existing graph hashing methods cannot achieve satisfactory performance in real applications due to the high complexity for graph modeling. In this paper, we propose a novel method, called scalable graph hashing with feature transformation (SGH), for large-scale graph hashing. Through feature transformation, we can effectively approximate the whole graph without explicitly computing the similarity graph matrix, based on which a sequential learning method is proposed to learn the hash functions in a bit-wise manner. Experiments on two datasets with one million data points show that our SGH method can outperform the state-of-the-art methods in terms of both accuracy and scalability.

Recent advances in imaging and multimedia technologies have paved the way for automatic analysis of visual art. Despite notable attempts, extracting relevant patterns from paintings is still a challenging task. Different painters, born in different periods and places, have been influenced by different schools of arts. However, each individual artist also has a unique signature, which is hard to detect with algorithms and objective features. In this paper we propose a novel dictionary learning approach to automatically uncover the artistic style from paintings. Specifically, we present a multi-task learning algorithm to learn a style-specific dictionary representation. Intuitively, our approach, by automatically decoupling style-specific and artist-specific patterns, is expected to be more accurate for retrieval and recognition tasks than generic methods. To demonstrate the effectiveness of our approach, we introduce the DART dataset, containing more than 1.5K images of paintings representative of different styles. Our extensive experimental evaluation shows that our approach significantly outperforms state-of-the-art methods.

In this paper, we propose a new local spatio-temporal descriptor for videos and we propose a new approach for action recognition in videos based on the introduced descriptor. The new descriptor is called the Video Covariance Matrix Logarithm (VCML). The VCML descriptor is based on a covariance matrix representation, and it models relationships between different low-level features, such as intensity and gradient. We apply the VCML descriptor to encode appearance information of local spatio-temporal video volumes, which are extracted by the Dense Trajectories. Then, we present an extensive evaluation of the proposed VCML descriptor with the Fisher vector encoding and the Support Vector Machines on four challenging action recognition datasets. We show that the VCML descriptor achieves better results than the state-of-the-art appearance descriptors. Moreover, we present that the VCML descriptor carries complementary information to the HOG descriptor and their fusion gives a significant improvement in action recognition accuracy. Finally, we show that the VCML descriptor improves action recognition accuracy in comparison to the state-of-the-art Dense Trajectories, and that the proposed approach achieves superior performance to the state-of-the-art methods.

This paper addresses the task of time separated aerial image registration. The ability to solve this problem accurately and reliably is important for a variety of subsequent image understanding applications. The principal challenge lies in the extent and nature of transient appearance variation that a land area can undergo, such as that caused by the change in illumination conditions, seasonal variations, or the occlusion by non-persistent objects (people, cars). Our work introduces several novelties: (i) unlike all previous work on aerial image registration, we approach the problem using a set-based paradigm; (ii) we show how local, pair-wise constraints can be used to enforce a globally good registration using a constraints graph structure; (iii) we show how a simple holistic representation derived from raw aerial images can be used as a basic building block of the constraints graph in a manner which achieves both high registration accuracy and speed. We demonstrate: (i) that the proposed method outperforms the state-of-the-art for pair-wise registration already, achieving greater accuracy and reliability, while at the same time reducing the computational cost of the task; and (ii) that the increase in the number of available images in a set consistently reduces the average registration error.

Community detection on social media is a classic and challenging task. In this paper, we study the problem of detecting communities by combining social relations and user generated content in social networks. We propose a nonnegative matrix tri-factorization (NMTF) based clustering framework with three types of graph regularization. The NMTF based clustering framework can combine the relations and content seamlessly and the graph regularization can capture user similarity, message similarity and user interaction explicitly. In order to design regularization components, we further exploit user similarity and message similarity in social networks. A unified optimization problem is proposed by integrating the NMTF framework and the graph regularization. Then we derive an iterative learning algorithm for this optimization problem. Extensive experiments are conducted on three real-world data sets and the experimental results demonstrate the effectiveness of the proposed method.

We introduce the study of adversarial effects on wisdom of the crowd phenomena. In particular, we examine the ability of an adversary to influence a social network so that the majority of nodes are convinced by a falsehood, using its power to influence a certain fraction, μ < 0.5 of N experts. Can a bad restaurant make a majority of the overall network believe in the quality of that restaurant by misleading a certain share of food critics into believing its food is good, and use the influence of those experts to make a majority of the overall network to believe in the quality of that restaurant? We are interested in providing an agent, which does not necessarily know the graph structure nor who the experts are, to determine the true value of a binary property using a simple majority. We prove bounds on the social graph's maximal degree, which ensure that with a high probability the adversary will fail (and the majority vote will coincide with the true value) when it can choose who the experts are, while each expert communicates the true value with probability p > 0.5. When we examine expander graphs as well as random graphs we prove such bounds even for stronger adversaries, who are able to pick and choose not only who the experts are, but also which ones of them would communicate the wrong values, as long as their proportion is 1-p. Furthermore, we study different propagation models and their effects on the feasibility of obtaining the true value for different adversary types.

They pinpoint families of voting rules that exhibit robustness: they are accurate in the limit with respect to a wide Classic social choice theory assumes that votes are range of noise models, which govern the way noisy votes are independent (but possibly conditioned on an underlying generated, given the ground truth [Caragiannis et al., 2013; objective ground truth). This assumption 2014]. is unrealistic in settings where the voters are connected While these results are promising, they rely on a crucial via an underlying social network structure, modeling assumption: votes are independent. This assumption as social interactions lead to correlated votes. We is clearly satisfied in some settings -- when votes are establish a general framework -- based on random submitted by computer Go programs [Jiang et al., 2014], say.

Many hard computational social choice problems are known to become tractable when voters' preferences belong to a restricted domain, such as those of single-peaked or single-crossing preferences. However, to date, all algorithmic results of this type have been obtained for the setting where each voter's preference list is a total order of candidates. The goal of this paper is to extend this line of research to the setting where voters' preferences are dichotomous, i.e., each voter approves a subset of candidates and disapproves the remaining candidates. We propose several analogues of the notions of single-peaked and single-crossing preferences for dichotomous profiles and investigate the relationships among them. We then demonstrate that for some of these notions the respective restricted domains admit efficient algorithms for computationally hard approval-based multi-winner rules.