We have developed a novel hybrid representation for Music Information Retrieval. Our representation is built by incorporating audio content into the tag space in a tag-track matrix, and then learning hybrid concepts using latent semantic analysis. We apply this representation to the task of music recommendation, using similarity-based retrieval from a query music track. We also develop a new approach to evaluating music recommender systems, which is based upon the relationship of users liking tracks. We are interested in measuring the recommendation quality, and the rate at which cold-start tracks are recommended. Our hybrid representation is able to outperform a tag-only representation, in terms of both recommendation quality and the rate that cold-start tracks are included as recommendations.

Given the number of online sources for news, the volumes of news generated are so daunting that gaining insight from these collections become impossible without some aid to link them. Semantic linking of news articles facilitates grouping of similar or relevant news stories together for ease of human consumption. For example, a political analyst may like to have a single view of all news articles that report visits of State heads of different countries to a single country to make an in-depth analytical report on the possible impacts of all associated events. It is likely that no news source links all the relevant news together. In this paper, we discuss a multi-resolution, multi-perspective news analysis system that can link news articles collected from diverse sources over a period of time. The distinctive feature of the proposed news linking system is its capability to simultaneously link news articles and stories at multiple levels of granularity. At the lowest level several articles reporting the same event are linked together. Higher level groupings are more contextual and semantic. We have deployed a range of algorithms that use statistical text processing and Natural Language Processing techniques. The system is incremental in nature and depicts how stories have evolved over time along with main actors and activities. It also illustrates how a single story diverges into multiple themes or multiple stories converge due to conceptual similarity. Accuracy of linking thematically and conceptually linked news articles are also presented.

The process of extracting useful knowledge from large datasets has become one of the most pressing problems in today’s society. The problem spans entire sectors, from scientists to intelligence analysts and web users, all of whom are constantly struggling to keep up with the larger and larger amounts of content published every day. With this much data, it is often easy to miss the big picture. In this paper, we investigate methods for automatically connecting the dots – providing a structured, easy way to navigate within a new topic and discover hidden connections. We focus on the news domain: given two news articles, our system automatically finds a coherent chain linking them together. For example, it can recover the chain of events leading from the decline of home prices (2007) to the health-care debate (2009). We formalize the characteristics of a good chain and provide efficient algorithms to connect two fixed endpoints. We incorporate user feedback into our framework, allowing the stories to be refined and personalized. Finally, we evaluate our algorithm over real news data. Our user studies demonstrate the algorithm's effectiveness in helping users understanding the news.

The personalization aspect of recommender systems makes them well suited to applications in The goal of rating-based recommender systems is electronic commerce and entertainment, while the fact that to make personalized predictions and recommendations they do not rely on text-based descriptions of items makes for individual users by leveraging the preferences them well suited to content like movies and music. of a community of users with respect to a In this paper, we focus on a key problem in rating-based collection of items like songs or movies. Recommender collaborative filtering: the possibility of a basic incompatibility systems are often based on intricate statistical between the properties of recommender system data sets models that are estimated from data sets containing and the assumptions required for valid estimation and evaluation a very high proportion of missing ratings. of statistical models in the presence of missing data. This work describes evidence of a basic incompatibility We describe properties of recommender system data sets and between the properties of recommender relate them to the statistical theory of model estimation in system data sets and the assumptions required for the presence of nonrandom missing data. We describe an valid estimation and evaluation of statistical models extended modelling framework and a modified set of evaluation in the presence of missing data. We discuss the protocols for dealing with nonrandom missing data.

Many systems use Markov models to generate finite-length sequences that imitate a given style. These systems often need to enforce specific control constraints on the sequences to generate. Unfortunately, control constraints are not compatible with Markov models, as they induce long-range dependencies that violate the Markov hypothesis of limited memory. Attempts to solve this issue using heuristic search do not give any guarantee on the nature and probability of the sequences generated. We propose a novel and efficient approach to controlled Markov generation for a specific class of control constraints that 1) guarantees that generated sequences satisfy control constraints and 2) follow the statistical distribution of the initial Markov model. Revisiting Markov generation in the framework of constraint satisfaction, we show how constraints can be compiled into a non-homogeneous Markov model, using arc-consistency techniques and renormalization. We illustrate the approach on a melody generation problem and sketch some realtime applications in which control constraints are given by gesture controllers.

There are two key issues for information diffusion in blogosphere: (1) blog posts are usually short, noisy and contain multiple themes, (2) information diffusion through blogosphere is primarily driven by the “word-of-mouth” effect, thus making topics evolve very fast. This paper presents a novel topic tracking approach to deal with these issues by modeling a topic as a semantic graph in which the semantic relatedness between terms are learned from Wikipedia. For a given topic/post, the named entities, Wikipedia concepts, and the semantic relatedness are extracted to generate the graph model. Noises are filtered out through a graph clustering algorithm. To handle topic evolution, the topic model is enriched by using Wikipedia as background knowledge. Furthermore, graph edit distance is used to measure the similarity between a topic and its posts. The proposed method is tested using real-world blog data. Experimental results show the advantage of the proposed method on tracking topics in short, noisy text.

Another example is items to users based on their historical ratings. In that, although many people don't like animations, they may real-world scenarios, user interests may drift over still have interests in emerging 3-D animations because of the time since they are affected by moods, contexts, fantastic 3-D visual effects. These observations show that, and pop culture trends. This leads to the fact that although many aspects of user interests can be found based a user's historical ratings comprise many aspects of on users' historical ratings, at a certain time slice, one user's user interests spanning a long time period. However, interest may only focus on one or a couple of aspects. Thus, at a certain time slice, one user's interest may the static CF methods built on the entire historical ratings are only focus on one or a couple of aspects. Thus, inadequate to capture user-interest drift. In order to track user CF techniques based on the entire historical ratings interests and create comprehensive user profiles such that different may recommend inappropriate items. In this paper, recommendation strategies can be used for consistenttaste we consider modeling user-interest drift over time users and changing-taste users, a CF method that can based on the assumption that each user has multiple model user interests over time is required.

Extracting the relations that exist between two entities is an important step in numerousWeb-related tasks such as information extraction.A supervised relation extraction system that is trained to extract a particular relation type might not accurately extract a new type of a relation for which it has not been trained.However, it is costly to create training data manually for every new relation type that one might want to extract.We propose a method to adapt an existing relation extraction system to extractnew relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.

Information extraction systems have been recently proposed for organizing and exploring content in large online text corpora as information networks . In such networks, the nodes are named entities (e.g., people, organizations) while the edges correspond to statements indicating relations among such entities. To date, such systems extract rather primitive networks, capturing only those relations which are expressed by direct statements. In many applications, it is useful to also extract more subtle relations which are often expressed as meta statements in the text. These can, for instance provide the context for a statement (e.g., “Google acquired YouTube on October 2006”), or repercussion about a statement (e.g., “The US condemned Russia’s invasion of Georgia”). In this work, we report on a system for extracting relations expressed in both direct statements as well as in meta statements. We propose a method based on Conditional Random Fields that explores syntactic features to extract both kinds of statements seamlessly. We follow the Open Information Extraction paradigm, where a classifier is trained to recognize any type of relation instead of specific ones. Finally, our results show substantial improvements over a state-of-the-art information extraction system, both in terms of accuracy and, especially, recall.

The increased popularity of feature-rich mobile devices in recent years has enabled widespread consumption and production of social media content via mobile devices. Because mobile devices and mobile applications change context within which an individual generates and consumes microblog content, we might expect microblogging behavior to differ depending on whether the user is using a mobile device. To our knowledge, little has been established about what, if any, effects such mobile interfaces have on microblogging. In this paper, we investigate this question within the context of Twitter, among the most popular microblogging platforms. This work makes three specific contributions. First, we quantify the ways in which user profiles are effected by the mobile context: (1) the extent to which users tend to be either fully non-mobile or mobile and (2) the relative activity of the mobile Twitter community. Second, we assess the differences in content between mobile and non-mobile tweets (posts to the Twitter platform). Our results show that mobile platforms produce very different patterns of Twitter usage. As part of our analysis, we propose and apply a classification system for tweets. We consider this to be the third contribution of this work. While other classification systems have been proposed, ours is the first to permit the independent encoding of a tweet’s form, content, and intended audience. In this paper we apply this system to show how tweets differ between mobile and non-mobile contexts. However, because of its flexibility and breadth, the schema may be useful to researchers studying Twitter content in other contexts as well.