Academic literature on the topic 'Query suggestions'

With the growing amount of XML and RDF data, keyword search over XML and RDF data has become an important and increasingly researched topic. In this thesis, we investigate several problems related to keyword search over XML and RDF data, and provide solutions to these problems. The research consists largely of three main technical contributions: top-k nearest keyword (NK) search for XML data, query suggestion for XML data and query suggestion for RDF data. We first study the top-k NK search problem for XML data, which provides an approach to exploring XML queries by the distance between keyword matching nodes. A top-k NK query is to find the top-k nearest neighbors of a given node where each neighbor matches a certain keyword, and it can serve as the building block to deal with many problems in XML data, such as keyword search. In our research, we design a method to construct an extended compact TVP (ecTVP) index to efficiently find the top-k nearest neighbors in XML data. We build the index by constructing a variant of Extended Compact Tree, which finds the top-k nearest neighbors during a bottom-up and a top-down process. Theoretical analysis and experiments indicate that our proposed method is efficient to build the ecTVP index. Moreover, we reduce the redundancy in the ecTVP index. In this way, the index costs less space and query time.<br>Thesis (PhD Doctorate)<br>Doctor of Philosophy (PhD)<br>School of Information and Communication Technology<br>Science, Environment, Engineering and Technology<br>Full Text

Current web search engines, such as Google, Yahoo!, and Bing, rank the set of documents S retrieved in response to a user query Q and display each document with a title and a snippet, which serves as an abstract of the corresponding document in S. Snippets, however, are not as useful as they are designed for, i.e., to assist search engine users to quickly identify results of interest, if they exist, without browsing through the documents in S, since they (i) often include very similar information and (ii) do not capture the main content of the corresponding documents. Moreover, when the intended information need specified in a search query is ambiguous, it is difficult, if not impossible, for a search engine to identify precisely the set of documents that satisfy the user's intended request. Furthermore, a document title retrieved by web search engines is not always a good indicator of the content of the corresponding document, since it is not always informative. All these design problems can be solved by our proposed query-based, web informative summarization engine, denoted Q-WISE. Q-WISE clusters documents in S, which allows users to view segregated document collections created according to the specific topic covered in each collection, and generates a concise/comprehensive summary for each collection/cluster of documents. Q-WISE is also equipped with a query suggestion module that provides a guide to its users in formulating a keyword query, which facilitates the web search and improves the precision and recall of the search results. Experimental results show that Q-WISE is highly effective and efficient in generating a high quality summary for each cluster of documents on a specific topic, retrieved in response to a Q-WISE user's query. The empirical study also shows that Q-WISE's clustering algorithm is highly accurate, labels generated for the clusters are useful and often reflect the topic of the corresponding clustered documents, and the performance of the query suggestion module of Q-WISE is comparable to commercial web search engines.

Query spelling correction is an important component of modern search engines that can help users to express their intent, and thus improve search quality. In this study, we investigated with what accuracy a sequence-to-sequence recurrent neural network (RNN) can recognise and correct misspellings in a music search engine, when the model is trained with old search queries. A sequence-to-sequence RNN was chosen as the model in this study since it has achieved state-of-the-art performance on similar tasks, such as machine translation and speech recognition. The findings from the study imply that the model learns to correct and complete queries with higher accuracy compared to a baseline model that returns the input query. However, we suggest that, for a model that would be good enough for production, more work needs to be done. Especially, work on creating a cleaner, less biased training dataset. Nevertheless, our work strengthens the idea that sequence-to-sequence RNNs could be used as a spell correction system in search engines.<br>Stavningskorrigering av söksträngar är en viktig komponent i moderna sökmotorer. Stavningskorrigering kan hjälpa användarna att uttrycka sig och därmed förbättra kvaliteten i sökningen. I det här arbetet undersökte vi med vilken noggrannhet en Recurrent neural network (RNN) modell kan lära sig att korrigera felstavningar i söksträngar från en sökmotor för musik. RNN modellen tränades med söksträngar från historiska sökningar från sökmotorn. Anledningen till att RNN valdes som modell i den här studien var för att den har uppnått hittills bästa möjliga resultat på liknande uppgifter, såsom maskinöversättning och taligenkänning. Resultaten från vår studie visar att modellen lär sig att korrigera och komplettera söksträngar med högre noggrannhet än en basmodell som enbart returnerar indatasträngen. För att utveckla en modell som är tillräckligt bra för produktion föreslår vi emellertid att mer arbete måste utföras. Framför allt är vi övertygade om att ett renare, mindre systematiskt avvikande träningsdataset skulle förbättra modellen. På det hela taget stärker dock vårt arbete hypothesen att RNN modeller kan användas som stavningskorrigeringssystem i sökmotorer.

In order to improve a user's query and help the user quickly satisfy his/her information need, most search engines provide query suggestions that are meant to be relevant alternatives to the user's query. This thesis builds on the query suggestion system and evaluation methodology described in Shen Jiang's Masters thesis (2008). Jiang's system constructs query suggestions by searching for lexical aliases of web documents and then applying query search to the lexical aliases. A lexical alias for a web document is a list of terms that return the web document in a top-ranked position. Query search is a search process that finds useful combinations of search terms. The main focus of this thesis is to supply alternatives for the components of Jiang's system. We suggest three term scoring mechanisms and generalize Jiang's lexical alias search to be a general search for terms that are useful for constructing good query suggestions. We also replace Jiang's top-down query search by a bottom-up beam search method. We experimentally show that our query suggestion method improves Jiang's system by 30% for short queries and 90% for long queries using Jiang's evaluation method. In addition, we add new evidence supporting Jiang's conclusion that terms in the user's initial query terms are important to include in the query suggestions. In addition, we explore the usefulness of document clustering in creating query suggestions. Our experimental results are the opposite of what we expected: query suggestion based on clustering does not perform nearly as well, in terms of the "coverage" scores we are using for evaluation, as our best method that is not based on document clustering.

碩士<br>國立臺灣大學<br>資訊工程學研究所<br>100<br>Query suggestion is an important and widely studied research task in information retrieval. Most previous methods focus on providing suggestions relevant to the single query that the user just submits. Recently, the ‘context’ during the search process is taken into consideration, which includes the previous queries and click information in the same search session. Most of these methods make use of query dependency in the logs and provide suggested queries for particular context. However, a large scale of logs is needed for these methods. Besides, they are not suitable to explain the change of query reformulation behavior within the search process. In this work, we try to make use of user reformulation behavior and model its change during the search process. Our goal is to provide suggestions that are more suitable for different stages within the search process. Experimental results show that the two models used in our work, Variable Order Markov model and Linear Regression, are able to improve the performance of some existing method and provide suggestions expected by the user. In addition, we further analyze the logs and find some patterns related to reformulation behavior. These patters help to explain the abilities of our models.

碩士<br>國立臺灣師範大學<br>資訊工程學系<br>102<br>The goal of this thesis is to automatically suggest query keywords from the search results returned by the search engine in order to further filter the large amount of search results by using these query keywords as the specialized queries. A two-level query suggestion method, called the M_PhRank, is proposed. The first level suggestion aims to provide the query terms, which can cover search results as many as possible, and the query terms in the second level should have clear meaning and lower overlap between their covered objects. Firstly, the coverage over search results is computed as the novelty score of a word, which is used to select the topic terms in the first level suggestion. Secondly, the semantic scores of words are estimated by using the random walk algorithm on the co-occurrence graph of words. The query keywords consisting of 2-3 non-topic terms form the candidate subtopic terms, whose semantic scores are computed according to the semantic scores of their composing words. According to the given suggestion number, the number of subtopic terms under the topic-terms is decided proportional to the coverage of the topic terms. Finally, the hierarchical query suggestion structure is constructed by the topic terms in first level and their corresponding subtopic terms on the second level. The empirical experiment results show that the M_PhRank method performs better than the baseline method on providing more semantics specific terms and high coverage with limited overlap increasing. Moreover, according to user survey, the hierarchy of query keyword suggestions constructed by M_PhRank gets high satisfaction on query assistance.

碩士<br>國立臺灣大學<br>圖書資訊學研究所<br>98<br>It has been argued that traditional information retrieval evaluation is ill-equipped to address the need to validate the efficacy of today’s highly interactive systems, which require users’ active participation to be effective. To answer the challenge of interactive information retrieval evaluation, a novel methodology was applied to test the effectiveness of MAP (Multiple Access to PubMed), a metadata-guided search interface for PubMed bibliographic search. The most distinctive aspect of our methodology is to use real users searching for real search requests on real system, instead of using assigned tasks common in traditional IR evaluation. To control the impact of individual search requests on search performance, a repeated measure design was adopted where users&apos;&apos; search request served as its own control of variance. Comparisons of information behaviors between MAP and the regular PubMed interface were made. The purpose of the study is to examine whether interfaces and topic familiarity might interfere user’s search and term selecting behaviors. Some major findings are as follows. 1. The participants were found to input more diverse terms and make more submissions were made using MAP. There were also differences in term category selection between the two interfaces. 2. Topic familiarity was also shown to influence users&apos;&apos; query expansion behaviors. Differences were found between the categories from which terms were selected between MAP and PubMed. 3. MAP was shown to help users to uncover relevance document ranked much lower in the original dataset. 4. MAP is more effective in search situations where the users were less familiar with the topics.

博士<br>國立臺灣大學<br>資訊工程學研究所<br>97<br>In statistical natural language processing, resources used to compute the statistics are indispensable. Different kinds of corpora have made available and many language models have been experimented. One major issue behind the corpus-based approaches is: if corpora adopted can reflect the up-to-date usage. As we know, languages are live. New terms and phrases are used in daily life. How to capture the new usages is an important research topic. This thesis defines a novel web-based relatedness measure and explores snippets in various web domains as corpora. Mutual dependency score between two objects is calculated according to content information and frequent information of the two objects. The relatedness score of the two objects is defined as projecting the dependency score by a transfer function. Four transfer functions based on Poisson, Log-concave Power-concave and Gompertz function are considered. Three famous benchmark datasets, including WordSimilarity-353, Miller-Charles and Rubenstein-Goodenough, verify the four transfer functions. Named entities are common foci of searchers. We apply the dependency score to evaluate named level association by three strategies, direct association, association matrix and scalar association matrix. Modeling and naming general entity-entity relationships is challenging in construction of social networks. Given a seed denoting a person name, we utilize Google search engine, NER (Named Entity Recognizer) parser, and the web-based relatedness measure to construct an evolving social network. For each entity pair in the network, we apply Markov chain random process to extract potential categories defined in the ODP. Moreover, for labeling their relationships, we try to combine the tf×idf scores of noun phrases extracted from snippets and the rank scores of the categories. Different from traditional query suggestion which is extracted from query logs,we extract suggestion terms from snippets. We apply our relatedness measures to the query suggestion. Using the proposed relatedness measures, our query suggestion extracted shows a high agreement of relatedness.

La recherche d'informations s'intéresse, entre autres, à répondre à des questions comme: est-ce qu'un document est pertinent à une requête ? Est-ce que deux requêtes ou deux documents sont similaires ? Comment la similarité entre deux requêtes ou documents peut être utilisée pour améliorer l'estimation de la pertinence ? Pour donner réponse à ces questions, il est nécessaire d'associer chaque document et requête à des représentations interprétables par ordinateur. Une fois ces représentations estimées, la similarité peut correspondre, par exemple, à une distance ou une divergence qui opère dans l'espace de représentation. On admet généralement que la qualité d'une représentation a un impact direct sur l'erreur d'estimation par rapport à la vraie pertinence, jugée par un humain. Estimer de bonnes représentations des documents et des requêtes a longtemps été un problème central de la recherche d'informations. Le but de cette thèse est de proposer des nouvelles méthodes pour estimer les représentations des documents et des requêtes, la relation de pertinence entre eux et ainsi modestement avancer l'état de l'art du domaine. Nous présentons quatre articles publiés dans des conférences internationales et un article publié dans un forum d'évaluation. Les deux premiers articles concernent des méthodes qui créent l'espace de représentation selon une connaissance à priori sur les caractéristiques qui sont importantes pour la tâche à accomplir. Ceux-ci nous amènent à présenter un nouveau modèle de recherche d'informations qui diffère des modèles existants sur le plan théorique et de l'efficacité expérimentale. Les deux derniers articles marquent un changement fondamental dans l'approche de construction des représentations. Ils bénéficient notamment de l'intérêt de recherche dont les techniques d'apprentissage profond par réseaux de neurones, ou deep learning, ont fait récemment l'objet. Ces modèles d'apprentissage élicitent automatiquement les caractéristiques importantes pour la tâche demandée à partir d'une quantité importante de données. Nous nous intéressons à la modélisation des relations sémantiques entre documents et requêtes ainsi qu'entre deux ou plusieurs requêtes. Ces derniers articles marquent les premières applications de l'apprentissage de représentations par réseaux de neurones à la recherche d'informations. Les modèles proposés ont aussi produit une performance améliorée sur des collections de test standard. Nos travaux nous mènent à la conclusion générale suivante: la performance en recherche d'informations pourrait drastiquement être améliorée en se basant sur les approches d'apprentissage de représentations.<br>Information retrieval is generally concerned with answering questions such as: is this document relevant to this query? How similar are two queries or two documents? How query and document similarity can be used to enhance relevance estimation? In order to answer these questions, it is necessary to access computational representations of documents and queries. For example, similarities between documents and queries may correspond to a distance or a divergence defined on the representation space. It is generally assumed that the quality of the representation has a direct impact on the bias with respect to the true similarity, estimated by means of human intervention. Building useful representations for documents and queries has always been central to information retrieval research. The goal of this thesis is to provide new ways of estimating such representations and the relevance relationship between them. We present four articles that have been published in international conferences and one published in an information retrieval evaluation forum. The first two articles can be categorized as feature engineering approaches, which transduce a priori knowledge about the domain into the features of the representation. We present a novel retrieval model that compares favorably to existing models in terms of both theoretical originality and experimental effectiveness. The remaining two articles mark a significant change in our vision and originate from the widespread interest in deep learning research that took place during the time they were written. Therefore, they naturally belong to the category of representation learning approaches, also known as feature learning. Differently from previous approaches, the learning model discovers alone the most important features for the task at hand, given a considerable amount of labeled data. We propose to model the semantic relationships between documents and queries and between queries themselves. The models presented have also shown improved effectiveness on standard test collections. These last articles are amongst the first applications of representation learning with neural networks for information retrieval. This series of research leads to the following observation: future improvements of information retrieval effectiveness has to rely on representation learning techniques instead of manually defining the representation space.