Matching Items (5)
Description
Query Expansion is a functionality of search engines that suggest a set of related queries for a user issued keyword query. In case of exploratory or ambiguous keyword queries, the main goal of the user would be to identify and select a specific category of query results among different categorical options, in order to narrow down the search and reach the desired result. Typical corpus-driven keyword query expansion approaches return popular words in the results as expanded queries. These empirical methods fail to cover all semantics of categories present in the query results. More importantly these methods do not consider the semantic relationship between the keywords featured in an expanded query. Contrary to a normal keyword search setting, these factors are non-trivial in an exploratory and ambiguous query setting where the user's precise discernment of different categories present in the query results is more important for making subsequent search decisions. In this thesis, I propose a new framework for keyword query expansion: generating a set of queries that correspond to the categorization of original query results, which is referred as Categorizing query expansion. Two approaches of algorithms are proposed, one that performs clustering as pre-processing step and then generates categorizing expanded queries based on the clusters. The other category of algorithms handle the case of generating quality expanded queries in the presence of imperfect clusters.
ContributorsNatarajan, Sivaramakrishnan (Author) / Chen, Yi (Thesis advisor) / Candan, Selcuk (Committee member) / Sen, Arunabha (Committee member) / Arizona State University (Publisher)
Created2011
Description
As pointed out in the keynote speech by H. V. Jagadish in SIGMOD'07, and also commonly agreed in the database community, the usability of structured data by casual users is as important as the data management systems' functionalities. A major hardness of using structured data is the problem of easily retrieving information from them given a user's information needs. Learning and using a structured query language (e.g., SQL and XQuery) is overwhelmingly burdensome for most users, as not only are these languages sophisticated, but the users need to know the data schema. Keyword search provides us with opportunities to conveniently access structured data and potentially significantly enhances the usability of structured data. However, processing keyword search on structured data is challenging due to various types of ambiguities such as structural ambiguity (keyword queries have no structure), keyword ambiguity (the keywords may not be accurate), user preference ambiguity (the user may have implicit preferences that are not indicated in the query), as well as the efficiency challenges due to large search space. This dissertation performs an expansive study on keyword search processing techniques as a gateway for users to access structured data and retrieve desired information. The key issues addressed include: (1) Resolving structural ambiguities in keyword queries by generating meaningful query results, which involves identifying relevant keyword matches, identifying return information, composing query results based on relevant matches and return information. (2) Resolving structural, keyword and user preference ambiguities through result analysis, including snippet generation, result differentiation, result clustering, result summarization/query expansion, etc. (3) Resolving the efficiency challenge in processing keyword search on structured data by utilizing and efficiently maintaining materialized views. These works deliver significant technical contributions towards building a full-fledged search engine for structured data.
ContributorsLiu, Ziyang (Author) / Chen, Yi (Thesis advisor) / Candan, Kasim S (Committee member) / Davulcu, Hasan (Committee member) / Jagadish, H V (Committee member) / Arizona State University (Publisher)
Created2011
Description
Speech recognition and keyword detection are becoming increasingly popular applications for mobile systems. While deep neural network (DNN) implementation of these systems have very good performance,
they have large memory and compute resource requirements, making their implementation on a mobile device quite challenging. In this thesis, techniques to reduce the memory and computation cost
of keyword detection and speech recognition networks (or DNNs) are presented.
The first technique is based on representing all weights and biases by a small number of bits and mapping all nodal computations into fixed-point ones with minimal degradation in the
accuracy. Experiments conducted on the Resource Management (RM) database show that for the keyword detection neural network, representing the weights by 5 bits results in a 6 fold reduction in memory compared to a floating point implementation with very little loss in performance. Similarly, for the speech recognition neural network, representing the weights by 6 bits results in a 5 fold reduction in memory while maintaining an error rate similar to a floating point implementation. Additional reduction in memory is achieved by a technique called weight pruning,
where the weights are classified as sensitive and insensitive and the sensitive weights are represented with higher precision. A combination of these two techniques helps reduce the memory
footprint by 81 - 84% for speech recognition and keyword detection networks respectively.
Further reduction in memory size is achieved by judiciously dropping connections for large blocks of weights. The corresponding technique, termed coarse-grain sparsification, introduces
hardware-aware sparsity during DNN training, which leads to efficient weight memory compression and significant reduction in the number of computations during classification without
loss of accuracy. Keyword detection and speech recognition DNNs trained with 75% of the weights dropped and classified with 5-6 bit weight precision effectively reduced the weight memory
requirement by ~95% compared to a fully-connected network with double precision, while showing similar performance in keyword detection accuracy and word error rate.
they have large memory and compute resource requirements, making their implementation on a mobile device quite challenging. In this thesis, techniques to reduce the memory and computation cost
of keyword detection and speech recognition networks (or DNNs) are presented.
The first technique is based on representing all weights and biases by a small number of bits and mapping all nodal computations into fixed-point ones with minimal degradation in the
accuracy. Experiments conducted on the Resource Management (RM) database show that for the keyword detection neural network, representing the weights by 5 bits results in a 6 fold reduction in memory compared to a floating point implementation with very little loss in performance. Similarly, for the speech recognition neural network, representing the weights by 6 bits results in a 5 fold reduction in memory while maintaining an error rate similar to a floating point implementation. Additional reduction in memory is achieved by a technique called weight pruning,
where the weights are classified as sensitive and insensitive and the sensitive weights are represented with higher precision. A combination of these two techniques helps reduce the memory
footprint by 81 - 84% for speech recognition and keyword detection networks respectively.
Further reduction in memory size is achieved by judiciously dropping connections for large blocks of weights. The corresponding technique, termed coarse-grain sparsification, introduces
hardware-aware sparsity during DNN training, which leads to efficient weight memory compression and significant reduction in the number of computations during classification without
loss of accuracy. Keyword detection and speech recognition DNNs trained with 75% of the weights dropped and classified with 5-6 bit weight precision effectively reduced the weight memory
requirement by ~95% compared to a fully-connected network with double precision, while showing similar performance in keyword detection accuracy and word error rate.
ContributorsArunachalam, Sairam (Author) / Chakrabarti, Chaitali (Thesis advisor) / Seo, Jae-Sun (Thesis advisor) / Cao, Yu (Committee member) / Arizona State University (Publisher)
Created2016
Description
Keyword search provides a simple and user-friendly mechanism for information search, and has become increasingly popular for accessing structured or semi-structured data. However, there are two open issues of keyword search on semi/structured data which are not well addressed by existing work yet.
First, while an increasing amount of investigation has been done in this important area, most existing work concentrates on efficiency instead of search quality and may fail to deliver high quality results from semantic perspectives. Majority of the existing work generates minimal sub-graph results that are oblivious to the entity and relationship semantics embedded in the data and in the user query. There are also studies that define results to be subtrees or subgraphs that contain all query keywords but are not necessarily ``minimal''. However, such result construction method suffers from the same problem of semantic mis-alignment between data and user query. In this work the semantics of how to {\em define} results that can capture users' search intention and then the generation of search intention aware results is studied.
Second, most existing research is incapable of handling large-scale structured data. However, as data volume has seen rapid growth in recent years, the problem of how to efficiently process keyword queries on large-scale structured data becomes important. MapReduce is widely acknowledged as an effective programming model to process big data. For keyword query processing on data graph, first graph algorithms which can efficiently return query results that are consistent with users' search intention are proposed. Then these algorithms are migrated to MapReduce to support big data. For keyword query processing on schema graph, it first transforms a keyword query into multiple SQL queries, then all generated SQL queries are run on the structured data. Therefore it is crucial to find the optimal way to execute a SQL query using MapReduce, which can minimize the processing time. In this work, a system called SOSQL is developed which generates the optimal query execution plan using MapReduce for a SQL query $Q$ with time complexity $O(n^2)$, where $n$ is the number of input tables of $Q$.
First, while an increasing amount of investigation has been done in this important area, most existing work concentrates on efficiency instead of search quality and may fail to deliver high quality results from semantic perspectives. Majority of the existing work generates minimal sub-graph results that are oblivious to the entity and relationship semantics embedded in the data and in the user query. There are also studies that define results to be subtrees or subgraphs that contain all query keywords but are not necessarily ``minimal''. However, such result construction method suffers from the same problem of semantic mis-alignment between data and user query. In this work the semantics of how to {\em define} results that can capture users' search intention and then the generation of search intention aware results is studied.
Second, most existing research is incapable of handling large-scale structured data. However, as data volume has seen rapid growth in recent years, the problem of how to efficiently process keyword queries on large-scale structured data becomes important. MapReduce is widely acknowledged as an effective programming model to process big data. For keyword query processing on data graph, first graph algorithms which can efficiently return query results that are consistent with users' search intention are proposed. Then these algorithms are migrated to MapReduce to support big data. For keyword query processing on schema graph, it first transforms a keyword query into multiple SQL queries, then all generated SQL queries are run on the structured data. Therefore it is crucial to find the optimal way to execute a SQL query using MapReduce, which can minimize the processing time. In this work, a system called SOSQL is developed which generates the optimal query execution plan using MapReduce for a SQL query $Q$ with time complexity $O(n^2)$, where $n$ is the number of input tables of $Q$.
ContributorsShan, Yi (Author) / Chen, Yi (Thesis advisor) / Bansal, Srividya (Thesis advisor) / Liu, Huan (Committee member) / Davulcu, Hasan (Committee member) / Arizona State University (Publisher)
Created2016
Description
Text search is a very useful way of retrieving document information from a particular website. The public generally use internet search engines over the local enterprise search engines, because the enterprise content is not cross linked and does not follow a page rank algorithm. On the other hand the enterprise search engine uses metadata information, which allows the user to specify the conditions that any retrieved document should meet. Therefore, using metadata information for searching will also be very useful. My thesis aims on developing an enterprise search engine using metadata information by providing advanced features like faceted navigation. The search engine data was extracted from various Indonesian web sources. Metadata information like person, organization, location, and sentiment analytic keyword entities should be tagged in each document to provide facet search capability. A shallow parsing technique like named entity recognizer is used for this purpose. There are more than 1500 entities that have been tagged in this process. These documents have been successfully converted into XML format and are indexed with "Apache Solr". It is an open source enterprise search engine with full text search and faceted search capabilities. The entities will be helpful for users to specify conditions and search faster through the large collection of documents. The user is assured results by clicking on a metadata condition. Since the sentiment analytic keywords are tagged with positive and negative values, social scientists can use these results to check for overlapping or conflicting organizations and ideologies. In addition, this tool is the first of its kind for the Indonesian language. The results are fetched much faster and with better accuracy.
ContributorsSanaka, Srinivasa Raviteja (Author) / Davulcu, Hasan (Thesis advisor) / Sen, Arunabha (Committee member) / Taylor, Thomas (Committee member) / Arizona State University (Publisher)
Created2010