Query by document

Abstract

We are experiencing an unprecedented increase of content contributed by users in forums such as blogs, social networking sites and microblogging services. Such abundance of content complements content on web sites and traditional media forums such as news papers, news and financial streams, and so on. Given such plethora of information there is a pressing need to cross reference information across textual services. For example, commonly we read a news item and we wonder if there are any blogs reporting related content or vice versa.In this paper, we present techniques to automate the process of cross referencing online information content. We introduce methodologies to extract phrases from a given "query document" to be used as queries to search interfaces with the goal to retrieve content related to the query document. In particular, we consider two techniques to extract and score key phrases. We also consider techniques to complement extracted phrases with information present in external sources such as Wikipedia and introduce an algorithm called RelevanceRank for this purpose.We discuss both these techniques in detail and provide an experimental study utilizing a large number of human judges from Amazons's Mechanical Turk service. Detailed experiments demonstrate the effectiveness and efficiency of the proposed techniques for the task of automating retrieval of documents related to a query document.

Cross Domain Search by Exploiting Wikipedia

“…They first extract semantic keywords from the paragraph and then do the search in an annotated image database. In addition, querying by documents is studied in [24].…”

Monetizing User Activity on Social Networks - Challenges and Experiences

“…We chose YTE over frequency based techniques since we did not want to be limited by counts from a 12000 post corpus for tf.idf calculations. Also, a recent work comparing YTE, tf.idf and mutual information based techniques for word and phrase identification concluded that YTE did better than tf.idf when identifying top k < 4 keywords in a document and all three were similar in characterizing document content for larger values of k [6].…”

“…As every keyword k i is added from C2 to C1, the change in Information Content of C1 is measured as IC(C1,ki ) δ =IC(C1,ki )−IC(C1) (6) where IC(C1, k i ) is the information content of C1 after adding keyword k i from C2. IC(C1, k i ) δ is positive when k i is strongly associated with words in C1 and negative when k i is unrelated to words in C1.…”

Learning Valued Relations from Data

“…Relational data can be observed in many predictive modeling tasks, such as forecasting the winner in two-player computer games [1], predicting proteins that interact with other proteins in bioinformatics [2], retrieving documents that are similar to a target document in text mining [3], investigating the persons that are friends of each other on social network sites [4], etc. All these examples represent fields of applica-tion in which specific machine learning and data mining algorithms are successfully developed to infer relations from data; pairwise relations, to be more specific.…”