The American Football betting market provides a particularly attractive domain to study the nexus between public sentiment and the wisdom of crowds. In this paper, we present the first substantial study of the relationship between the NFL betting line and public opinion expressed in blogs and microblogs (Twitter). We perform a large-scale study of four distinct text streams: LiveJournal blogs, RSS blog feeds captured by Spinn3r, Twitter, and traditional news media. Our results show interesting disparities between the first and second halves of each season. We present evidence showing usefulness of sentiment on NFL betting. We demonstrate that a strategy betting roughly 30 games per year identified winner roughly 60% of the time from 2006 to 2009, well beyond what is needed to overcome the bookie's typical commission(53%).

This paper discusses a research based on analyzing criminal sentences on criminal trials on organized crime activity in Sicily pronounced from 2000 through 2006. Large criminal sentences related dataset collection activity in Italy is severely constrained for various reasons such as difficulty of data collection at the courthouses, unavailability of data in digital format, and classification criteria used in the public archives. Thus, in general, judicial statistics suffer from lack of reliability and informativeness. The objective of this research is to analyze the text of criminal sentences in a revisable and verifiable way, so that information is extracted on the trial leading to the sentence, the socio-economic environment in which the relevant events occurred, and the differences between the various districts conducting the trials. The purpose is to elaborate a tool of automated analysis of the text of the sentences that is generalizable to other areas of jurisprudence, and, outside of jurisprudence, to other temporal and geographical contexts. The 726 criminal sentences that have been converted into text files have been pronounced at all judicial levels in the four Sicilian districts for mafia-related crimes. This research is relevant because, for the first time in Italy, we aim to empirically describe the juridical response to the phenomenon of organized crime, by using a large and extendable database of criminal sentences that can be analyzed with data mining techniques, rather than deriving general conclusions from a focused small set of sentences.

Textual sentiment classifiers classify texts into a fixed number of affective classes, such as positive, negative or neutral sentiment, or subjective versus objective information. It has been observed that sentiment classifiers suffer from a lack of generalization capability: a classifier trained on a certain domain generally performs worse on data from another domain. This phenomenon has been attributed to domain-specific affective vocabulary. In this paper, we propose a voting-based thresholding approach, which calibrates a number of existing single-domain classifiers with respect to sentiment data from a new domain. The approach presupposes only a small amount of annotated data from the new domain. We evaluate three criteria for estimating thresholds, and discuss the ramifications of these criteria for the trade-off between classifier performance and manual annotation effort.

For many supervised learning problems, we possess prior knowledge about which features yield similar information about the target variable. In predicting the topic of a document, we might know that two words are synonyms, or when performing image recognition, we know which pixels are adjacent. Such synonymous or neighboring features are near-duplicates and should therefore be expected to have similar weights in a good model. Here we present a framework for regularized learning in settings where one has prior knowledge about which features are expected to have similar and dissimilar weights. This prior knowledge is encoded as a graph whose vertices represent features and whose edges represent similarities and dissimilarities between them. During learning, each feature's weight is penalized by the amount it differs from the average weight of its neighbors. For text classification, regularization using graphs of word co-occurrences outperforms manifold learning and compares favorably to other recently proposed semi-supervised learning methods. For sentiment analysis, feature graphs constructed from declarative human knowledge, as well as from auxiliary task learning, significantly improve prediction accuracy.

Automatic annotation of documents with controlled vocabulary terms (descriptors) from a conceptual thesaurus is not only useful for document indexing and retrieval. The mapping of texts onto the same thesaurus furthermore allows to establish links between similar documents. This is also a substantial requirement of the Semantic Web. This paper presents an almost language-independent system that maps documents written in different languages onto the same multilingual conceptual thesaurus, EUROVOC. Conceptual thesauri differ from Natural Language Thesauri in that they consist of relatively small controlled lists of words or phrases with a rather abstract meaning. To automatically identify which thesaurus descriptors describe the contents of a document best, we developed a statistical, associative system that is trained on texts that have previously been indexed manually. In addition to describing the large number of empirically optimised parameters of the fully functional application, we present the performance of the software according to a human evaluation by professional indexers.

Previous work on information extraction from unstructured, ungrammatical text (e.g. classified ads) showed that exploiting a set of background knowledge, called a "reference set," greatly improves the precision and recall of the extractions. However, finding a source for this reference set is often difficult, if not impossible. Further, even if a source is found, it might not overlap well with the text for extraction. In this paper we present an approach to building the reference set directly from the text itself. Our approach eliminates the need to find the source for the reference set, and ensures better overlap between the text and reference set. Starting with a small amount of background knowledge, our technique constructs tuples representing the entities in the text to form a reference set. Our results show that our method outperforms manually constructed reference sets, since hand built reference sets may not overlap with the entities in the unstructured, ungrammatical text. We also ran experiments comparing our method to the supervised approach of Conditional Random Fields (CRFs) using simple, generic features. These results show our method achieves an improvement in F1-measure for 6/9 attributes and is competitive in performance on the others, and this is without training data.

In most sentiment analysis applications, the sentiment lexicon plays a key role. However, it is hard, if not impossible, to collect and maintain a universal sentiment lexicon for all application domains because different words may be used in different domains. The main existing technique extracts such sentiment words from a large domain corpus based on different conjunctions and the idea of sentiment coherency in a sentence. In this paper, we propose a novel propagation approach that exploits the relations between sentiment words and topics or product features that the sentiment words modify, and also sentiment words and product features themselves to extract new sentiment words. As the method propagates information through both sentiment words and features, we call it double propagation. The extraction rules are designed based on relations described in dependency trees. A new method is also proposed to assign polarities to newly discovered sentiment words in a domain. Experimental results show that our approach is able to extract a large number of new sentiment words. The polarity assignment method is also effective.

In order for agents to act on behalf of users, they will have to retrieve and integrate vast amounts of textual data on the World Wide Web. However, much of the useful data on the Web is neither grammatical nor formally structured, making querying difficult. Examples of these types of data sources are online classifieds like Craigslist and auction item listings like eBay. We call this unstructured, ungrammatical data "posts." The unstructured nature of posts makes query and integration difficult because the attributes are embedded within the text. Also, these attributes do not conform to standardized values, which prevents queries based on a common attribute value. The schema is unknown and the values may vary dramatically making accurate search difficult. Creating relational data for easy querying requires that we define a schema for the embedded attributes and extract values from the posts while standardizing these values. Traditional information extraction (IE) is inadequate to perform this task because it relies on clues from the data, such as structure or natural language, neither of which are found in posts. Furthermore, traditional information extraction does not incorporate data cleaning, which is necessary to accurately query and integrate the source. The two-step approach described in this paper creates relational data sets from unstructured and ungrammatical text by addressing both issues. To do this, we require a set of known entities called a "reference set." The first step aligns each post to each member of each reference set. This allows our algorithm to define a schema over the post and include standard values for the attributes defined by this schema. The second step performs information extraction for the attributes, including attributes not easily represented by reference sets, such as a price. In this manner we create a relational structure over previously unstructured data, supporting deep and accurate queries over the data as well as standard values for integration. Our experimental results show that our technique matches the posts to the reference set accurately and efficiently and outperforms state-of-the-art extraction systems on the extraction task from posts.

MetLife processes over 260,000 life insurance applications a year. MetLife's intelligent text analyzer (MITA) uses the information-extraction technique of natural language processing to structure the extensive textual fields on a life insurance application. MITA is currently processing 20,000 life insurance applications a month. Eighty-nine percent of the textual fields processed by MITA exceed the established confidence-level threshold and are potentially available for further analysis by domain-specific analyzers.

MetLife processes over 260,000 life insurance applications a year. Underwriting of these applications is labor intensive. Automation is difficult because the applications include many free-form text fields. MetLife's intelligent text analyzer (MITA) uses the information-extraction technique of natural language processing to structure the extensive textual fields on a life insurance application. Knowledge engineering, with the help of underwriters as domain experts, was performed to elicit significant concepts for both medical and occupational textual fields. A corpus of 20,000 life insurance applications provided the syntactical and semantic patterns in which these underwriting concepts occur. These patterns, in conjunction with the concepts, formed the frameworks for information extraction. Extension of the information-extraction work developed by Wendy Lehnert was used to populate these frameworks with classes obtained from the systematized nomenclature of human and veterinary medicine and the Dictionary of Occupational Titles ontologies. These structured frameworks can then be analyzed by conventional knowledge-based systems. MITA is currently processing 20,000 life insurance applications a month. Eighty-nine percent of the textual fields processed by MITA exceed the established confidence-level threshold and are potentially available for further analysis by domain-specific analyzers.