Models such as latent semantic analysis and those based on neural embeddings learn distributed representations of text, and match the query against the document in the latent semantic space. In traditional information retrieval models, on the other hand, terms have discrete or local representations, and the relevance of a document is determined by the exact matches of query terms in the body text. We hypothesize that matching with distributed representations complements matching with traditional local representations, and that a combination of the two is favorable. We propose a novel document ranking model composed of two separate deep neural networks, one that matches the query and the document using a local representation, and another that matches the query and the document using learned distributed representations. The two networks are jointly trained as part of a single neural network. We show that this combination or 'duet' performs significantly better than either neural network individually on a Web page ranking task, and also significantly outperforms traditional baselines and other recently proposed models based on neural networks. 42 44 46 48 50

No abstract

No abstract

For the TREC 2004 Novelty track, UMass participated in all four tasks. Although finding relevant sentences was harder this year than last, we continue to show marked improvements over the baseline of calling all sentences relevant, with a variant of tfidf being the most successful approach. We achieve 5-9% improvements over the baseline in locating novel sentences, primarily by looking at the similarity of a sentence to earlier sentences and focusing on named entities.For the High Accuracy Retrieval from Documents (HARD) track, we investigated the use of clarification forms, fixed-and variable-length passage retrieval, and the use of metadata. Clarification form results indicate that passage level feedback can provide improvements comparable to user supplied related-text for document evaluation and outperforms related-text for passage evaluation. Document retrieval methods without a query expansion component show the most gains from related-text. We also found that displaying the top passages for feedback outperformed displaying centroid passages. Named entity feedback resulted in mixed performance. Our primary findings for passage retrieval are that document retrieval methods performed better than passage retrieval methods on the passage evaluation metric of binary preference at 12,000 characters, and that clarification forms improved passage retrieval for every retrieval method explored. We found no benefit to using variable-length passages over fixed-length passages for this corpus. Our use of geography and genre metadata resulted in no significant changes in retrieval performance.

Social media platforms provide active communication channels during mass convergence and emergency events such as disasters caused by natural hazards. As a result, first responders, decision makers, and the public can use this information to gain insight into the situation as it unfolds. In particular, many social media messages communicated during emergencies convey timely, actionable information. Processing social media messages to obtain such information, however, involves solving multiple challenges including: parsing brief and informal messages, handling information overload, and prioritizing different types of information found in messages. These challenges can be mapped to classical information processing operations such as filtering, classifying, ranking, aggregating, extracting, and summarizing. We survey the state of the art regarding computational methods to process social media messages and highlight both their contributions and shortcomings. In addition, we examine their particularities, and methodically examine a series of key subproblems ranging from the detection of events to the creation of actionable and useful summaries. Research thus far has, to a large extent, produced methods to extract situational awareness information from social media. In this survey, we cover these various approaches, and highlight their benefits and shortcomings. We conclude with research challenges that go beyond situational awareness, and begin to look at supporting decision making and coordinating emergency-response actions.