Search

With the popularization of various applications of the Internet of Things (IoT), a large number of smart devices are connected to the IoT to form an interconnected network. Because the amount of network data generated by IoT devices is huge, data owners prefer to outsource these data to the cloud for releasing them from the local storage burden. Meanwhile, in order to protect the privacy of the network data generated by IoT devices, data owners usually encrypt them before outsourcing them to the cloud. Phrase search encryption plays an important role in the IoT-based applications. To the best of our knowledge, all existing phrase search encryption schemes cannot work when user makes spelling errors. In addition, many existing phrase search encryption schemes build the secure index based on the relative position of keywords, which may lead to inaccurate search results. In order to address these problems, we propose an intelligent fuzzy phrase search scheme over encrypted network data for IoT. In this scheme, we construct a fuzzy keyword set for each distinct keyword, which can efficiently address the spelling mistakes. Moreover, we build the binary matrix for the document collection, which can quickly achieve multi-keyword search. In order to accurately determine whether the keywords in the query phrases are consecutive, we build the look-up table for each document. Experimental result shows that the proposed scheme achieves high efficiency and accuracy.

Idioms are multi-word expressions whose meaning cannot always be deduced from the literal meaning of constituent words. A key feature of idioms that is central to this paper is their peculiar mixture of fixedness and variability, which poses challenges for their retrieval from large corpora using traditional search approaches. These challenges hinder insights into idiom usage affecting users who are conducting linguistic research as well as those involved in language ed-ucation. To facilitate access to idioms examples taken from real-world contexts, we introduce an information retrieval system designed specifically for idioms. Given a search query that represents an idiom, typically in its canonical form, the system expands it automatically to account for the most common types of idiom variation including inflection, open slots, adjectival or adverbial modification, and passivisation. As a by-product of query expansion, other types of idiom varia-tion captured include derivation, compounding, negation, distribution across multiple clauses as well as other unforeseen types of variation. The system was implemented on top of Elasticsearch, an open-source, distributed, scalable, real-time search engine. Flexible retrieval of idioms is supported by a combination of linguistic pre-processing of the search queries, their translation into a set of query clauses written in a query language called Query DSL, and analysis, an indexing process that involves tokenisation and normalisation. Our system outperformed the phrase search in terms of recall and outperformed the keyword search in terms of precision. Out of the three, our approach was found to provide the best balance between precision and recall. By providing a fast and easy way of finding idioms in large corpora, our approach can facilitate further developments in fields such as linguistics, language education and natural language processing. Keywords: information retrieval; natural language processing; corpus linguistics; multi-word expressions; idioms

Visual media is a better way to deliver the information than the old way of "reading". For that reason with the wide propagation of multimedia websites, there are large video library’s archives, which came to be a main resource for humans. This research puts its eyes on the existing development in applying classical phrase search methods to a linked vocal transcript and after that it retrieves the video, this an easier way to search any visual media. This system has been implemented using JSP and Java language for searching the speech in the videos

As cloud computing becomes prevalent, more and more data owners are likely to outsource their data to a cloud server. However, to ensure privacy, the data should be encrypted before outsourcing. Symmetric searchable encryption allows users to retrieve keyword over encrypted data without decrypting the data. Many existing schemes that are based on symmetric searchable encryption only support single keyword search, conjunctive keywords search, multiple keywords search, or single phrase search. However, some schemes, i.e., static schemes, only search one phrase in a query request. In this paper, we propose a multi-phrase ranked search over encrypted cloud data, which also supports dynamic update operations, such as adding or deleting files. We used an inverted index to record the locations of keywords and to judge whether the phrase appears. This index can search for keywords efficiently. In order to rank the results and protect the privacy of relevance score, the relevance score evaluation model is used in searching process on client-side. Also, the special construction of the index makes the scheme dynamic. The data owner can update the cloud data at very little cost. Security analyses and extensive experiments were conducted to demonstrate the safety and efficiency of the proposed scheme.

Cloud computing has generated much interest in the research community in recent years for its many advantages, but has also raise security and privacy concerns. The storage and access of confidential documents have been identified as one of the central problems in the area. In particular, many researchers investigated solutions to search over encrypted documents stored on remote cloud servers. While many schemes have been proposed to perform conjunctive keyword search, less attention has been noted on more specialized searching techniques. In this paper, we present a phrase search technique based on Bloom filters that is significantly faster than existing solutions, with similar or better storage and communication cost. Our technique uses a series of $n$ n -gram filters to support the functionality. The scheme exhibits a trade-off between storage and false positive rate, and is adaptable to defend against inclusion-relation attacks. A design approach based on an application's target false positive rate is also described.

The symmetric searchable encryption (SSE) scheme not only guarantees the privacy and security of cloud data, but also enables users to achieve fast search. However, the functions of the existing SSE scheme are not very perfect, for example, the phrase search cannot be effectively implemented. In order to solve the problem, this paper proposes an SSE scheme that can effectively support phrase search by means of the scheme of Cash et al. and proves that the scheme satisfies IND-CKA2 under the selected keywords through security analysis, and the search efficiency of the scheme can reach sub-linear levels.

This paper presents a visualization framework that aids readers in understanding and analyzing the contents of medium-sized text collections that are typical for the opus of a single or few authors. We contribute several document-based visualization techniques to facilitate the exploration of the work of the German author Bazon Brock by depicting various aspects of its texts, such as the TextGenetics that shows the structure of the collection along with its chronology. The ConceptCircuit augments the TextGenetics with entities – persons and locations that were crucial to his work. All visualizations are sensitive to a wildcard-based phrase search that allows complex requests towards the author's work. Further development, as well as expert reviews and discussions with the author Bazon Brock, focused on the assessment and comparison of visualizations based on automatic topic extraction against ones that are based on expert knowledge.

Objective: To study the efficient phrase search technique on the basis of Bloom filters that is considerably faster compared to existing solutions. Methods/Statistical Analysis: Nowadays, keyword search is very popular technique to retrieve the data from clouds, in which the keyword can be processed over encrypted data and retrieve most of irrelevant results. Many authors investigate the different searching techniques to retrieve sensitive information from the clouds. But few of them only focus on conjunctive keyword search. Hence, in this paper we present An Efficient Phrase Search technique using Bloom filters that are very fast compared with existing techniques and our scheme reduce storage and computational overheads. Findings: To evaluate our scheme we collect the set of 1500 documents and these documents are pre-processed to avoid headers and footers to decrease the distorted in the statistics of the given dataset. Then we apply the Bloom filters technique to improve the performance in our scheme. Finally, our scheme achieves lower storage and computational overheads rather than existing techniques. Applications/Improvements: Our proposed technique explores a series of n-gram filters to uphold the functionality. The scheme manifests a trade-off between storage and false positive rate. Keywords: Bloom Filters, Cloud Computing, Multiple Keyword Search, Phrase Search

Cloud Computing has seen a dramatic rise in adoption in the past decade amid se- curity and privacy concerns. One area of consensus is that encryption is necessary, as anonymization techniques have been shown to be unreliable. However, the processing of encrypted data has proven to be difficult. Briefly, the goal is to maintain security over remotely stored and accessed data while achieving reasonable storage cost and perfor- mance. Search is the most basic and central functionality of a privacy-protected cloud storage system actively being investigated. Recent works have looked at enabling more specialized search functions. In this thesis, we explore the problem of searching and pro- cessing of sequential data. We propose three solutions targeting textual data, with em- phasis respectively on security, storage cost and performance. Our first solution achieves a high level of security with reduced communication, storage and computational cost by exploiting properties of natural languages. Our second solution achieves a minimal storage cost by taking advantage of the space efficiency of Bloom filters. Both propos- als were also first to enable non-keyword search in phrases. Using a subsequence-based solution, our final phrase search scheme is currently the fastest phrase search protocol in literature. We also show how sequential data search schemes can be extended to in- clude auditing with minimal additional cost. The solution is capable of achieving proof of retrievability with unbounded number of audits. A sample application which enables searching and computing over target values of encrypted XML files is also demonstrated. In terms of media, we describe an encrypted cloud media storage solution that simultane- ously protects user privacy and enables copyright verification, and is the first to achieve security against dishonest participants. We also describe a framework where practical scalable privacy-protected copyright detection can be performed. Finally, an application of sequence querying over generic data in the form of an Anti-Virus over encrypted cloud storage is demonstrated. A private scanning solution and a public Anti-Virus as a ser- vice solution are described, noting that the technique can be conceptualized as a generic pattern matching solution on encrypted data. We also include some directions on future work and unexplored applications.

Phrase search enables relevant documents retrieval of an exact phrase in many cloud-based IoT machine learning applications to safeguard information from leakage. Here the documents are checked by the data proprietors before being exported to the cloud, rendering the search process a difficult job. Multi-keyword search evolving searchable encryption algorithms unable to establish the position association of phrases on the remote server in a parsed phrase over encrypted data. In this paper, an efficient privacy-preserving phrase search is introduced to protect users’ search patterns; a probabilistic phrase recognition algorithm is also used. The proposed work demonstrates the security of real-world datasets and compared with existing multi-keyword search schemes to show the better matched documents with moderate overheads.

In recent years, various disciplines have engaged in efforts to increase research reproducibility including the adoption of replicable search methodologies. With the development of reporting checklists and guidelines for systematic reviews such as the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement, authors are expected to transparently report search strategies. Replicable search strategies are critical since the included studies will be screened for inclusion in some forms of evidence synthesis, which could have practice and policy implications. In cases where search strategies miss germane literature, studies are open to further criticism and can face difficulties in peer review. In some cases, search strategies that contain nonalphanumeric or special characters may not retrieve pertinent literature due to a search platform’s capacity for handling said characters. In this study, we explore issues with phrasing search strategies containing curly and straight quotation marks tested in 40 search platforms. The discovery platforms were tested using quotation characters and the absence of said characters to investigate platform behavior. Searches were categorized into one of five groups: CI (curly ignored), CA (curly acknowledged), CU (curly unclear), NP (no phrases), and UC (unsupported characters). The study found that 42.5 percent of platforms ignored curly quotation marks and interpreted the test term as a phrase, 30 percent of the bibliographic platforms acknowledged curly quotations and completed the phrase search, and one platform flagged curly quotation marks as an unsupported character.

When submitting queries to information retrieval (IR) systems, users often have the option of specifying which, if any, of the query terms are heavily dependent on each other and should be treated as a fixed phrase, for instance by placing them between quotes. In addition to such cases where users specify term dependence, automatic ways also exist for IR systems to detect dependent terms in queries. Most IR systems use both user and algorithmic approaches. It is not however clear whether and to what extent user-defined term dependence agrees with algorithmic estimates of term dependence, nor which of the two may fetch higher performance gains. Simply put, is it better to trust users or the system to detect term dependence in queries? To answer this question, we experiment with 101 crowdsourced search engine users and 334 queries (52 train and 282 test TREC queries) and we record 10 assessments per query. We find that (i) user assessments of term dependence differ significantly from algorithmic assessments of term dependence (their overlap is approximately 30%); (ii) there is little agreement among users about term dependence in queries, and this disagreement increases as queries become longer; (iii) the potential retrieval gain that can be fetched by treating term dependence (both user- and system-defined) over a bag of words baseline is reserved to a small subset (approxi-mately 8%) of the queries, and is much higher for low-depth than deep preci-sion measures. Points (ii) and (iii) constitute novel insights into term dependence.

Information search is a common yet important task in everyday work and life. It remains challenging how to help users search for information or things they don’t know how to express with words. Also, even when people know how to express, the cognitive cost required to retrieve the concepts and formulate the queries can be excessive. In this paper, we present Click-Search, a search user interface that allows people to indicate their search intents by merely selecting and cropping segments of pictures. The system automatically converts cropped image segments to keywords based on known semantic labels at the level of image pixels that are generated by crowdsourced image tagging. Through a user study, we showed that Click-Search could support object-finding activities effectively with a satisfactory user experience.

Existing work of processing keyword searches on graph data focuses on efficiency of result generation. However, being oblivious to user search intention, a query result may contain multiple instances of user search target, and multiple query results may contain information for the same instance of user search target. With the misalignment between query results and search targets, a ranking function is unable to effectively rank the instances of search targets. In this paper we propose the concept of target-aware query results driven by inferred user search intention. We leverage the Information Theory and develop a general probability model to infer search targets by analyzing return specifiers, modifiers, relatedness relationships, and query keywords' information gain. Then we propose two important properties for a target-aware result: atomicity and intactness. We develop techniques to efficiently generate target-aware results. Extensive experimental evaluation shows the effectiveness and efficiency of our approach.

In disease information retrieval, usually, users only know a top of basic information about the disease such as headache and high fever symptoms. Most search engines rely on keyword search and do n...

Many multimedia retrieval tasks are faced with increasingly large-scale datasets and variously changing preferences of users in each query. There are at least three distinctive contextual aspects comprised to form a set of preferences of a user at each query time: content, intention, and response time. A content preference refers to the low-level or semantic representations of the data that a user is interested in. An intention preference refers to how the content should be regarded as relevant. And a response time preference refers to the ability to control a reasonable wait time. This paper features the dynamic adaptability of a multimedia search system to the contexts of its users and proposes a multicontext-adaptive indexing and search system for video data. The main contribution is the integration of context-based query creation functions with high-performance search algorithms into a unified search system. The indexing method modifies inverted list data structure in order to construct disk-resident databases for large-scale data and efficiently enables a dynamic pruning search mechanism on those indices. We implement a frame-wise video navigation system as an application of the indexing and search system using the a 2.14 TB movie dataset. Experimental studies on this system show the effectiveness of the proposed pruning search method when dealing with dynamic contexts and its comparative high search performance.

Background/Objectives: Traditional keyword-based search engines retrieve web pages by matching the exact tokens or words in the user query with the tokens or words in the web documents. This approach has many drawbacks. Synonyms or terms similar to tokens or words in the user query are not taken into consideration to search web pages. The keyword based search engine gives equal importance to all keywords whereas when user will enter query, he may have different levels of importance for different keywords in his opinion. To get the correct relevant result, users may need to enter several synonyms on his own to get the desired information which may otherwise result into the omission of many valuable web pages. Another problem is of information overloading. The traditional keyword based search engines make it very tedious for end user to locate the really useful information from a huge list of search results. Existing web is dominated by keyword based Search Engines which does not provide an appropriate mechanism to classify and locate the relevant search results. This leads to wastage of precious time of end user if he does not know the key terms which are utilized to index preferred correct pages. To resolve the above mentioned issues that the users face, in this paper we have proposed search techniques to develop Ontology based semantic search engine. Methods/Statistical Analysis: Ontology based Semantic Search Engine for the tourism domain is developed that understands the meaning of the user query and relatively provides the direct, precise and relevant result. Not only the user entered keyword based pages would be returned but also the pages that are appropriate enough with the meaning of the user entered keyword were also be returned by using the Ontological synonym dataset developed by using WordNet. Findings: Firstly, the Ontology Synonym set is constructed using WordNet and then the ontology synonym set parser is used to map the user defined query with the query prototype. By comparing the Query Similarity for every prototype, the service/sub domain with maximum query similarity is identified and the respective service is invoked. Also if the similarity is 100% the extra keywords are also considered to provide the relevant and precise results to the end user. Meta-processor will provide meta information about the URL. Application/ Improvements: The proposed Ontology based semantic search engine in tourism domain is an enhanced model. This model can be used by search engines, tourism industry professional and customers in getting better results of the searches undertaken. After testing in real world, the improvements can be worked out.

Query-based information retrieval is an essential part of the web search engine. Many researchers have applied different types of web mining technologies to find more relevant information based on the keyword but are not able to know the correct meaning of the term (keyword) single, multiword or phrases. In this paper we address this problem of searching phrases. In this work the phrase searching process is three-fold as whole Phrase, Sequence of term in phrase and Mingle of the term in the phrase. Here the user enters a query as phrases that is passed to various search engines and retrieves the top ‘n’ list of web pages. Initially preprocessing is performed on the Sequence of Keyword in phrase and Mingle of the keyword in the phrase. Then feature extraction is done based on the web pages in the various search engines using term Frequency-Inverse Document Frequency method. Following the feature extraction, grouping of the top ‘n’ list of web pages from various search engines based on the parameter as title-based, snippet-based, content-based, address-based, link-based, uniform resource locator-based, and co-occurrence-based calculation is done using LBG clustering algorithm. Then identified the unique link from the above grouping of web pages from the various search engines using SVM classifier and assigned the rank value to the unique link web pages are done using proposed ranking algorithm. Finally it is observed from this experiment that precision, recall, f-measure, accuracy, speed, and error rate show significant improvement than the traditional search engines.

In the Dictionary-based String Matching (DSM) problem, a retrieval system has access to a source sequence and stores the position of a certain number of strings in a posting table. When a user inquires the position of a string, the retrieval system, instead of searching in the source sequence directly, relies on the the posting table to answer the query more efficiently. In this paper, the Statistical DSM problem is a proposed as a statistical and information-theoretic formulation of the classic DSM problem in which both the source and the query have a statistical description while the strings stored in the posting sequence are described as a code. Through this formulation, we are able to define the efficiency of the retrieval system as the average cost in answering a users' query in the limit of sufficiently long source sequence. This formulation is used to study the retrieval performance for the case in which (i) all the strings of a given length, referred to as k-grams , and (ii) prefix-free codes., Comment: 8 pages, 2 figures, submitted to IEEE ICASSP 2019

Cloud computing has become a major computing service platform in the computer science field. However it has drawbacks and challenges related to the privacy of data on cloud. Even though scholars researched over it, proper solutions have not yet been proposed that facilitate searching over encrypted data stored on cloud server. Data breach is also one major threat to cloud storage. In this project, we provide a phrase search scheme using bloom filter to search encrypted data from cloud server using multiple keywords. Besides, we also introduce the Proxy Re-encryption scheme that enables data owner to provide access rights to their desired data users. This scheme assures high speed searching and faster response time at a low cost of communication.

Purpose Effective literature searches are critical to researchers and health care professionals. To conduct literature searches, clinicians, researchers and nurses rely primarily on a few major databases (PubMed, Cochrane, CINAHL, etc.) to retrieve information. However, there is a lack of literature on the comparative efficiencies of major databases for systematic review results on a clinically related topic. This paper aims to fill that gap in the literature. Design/methodology/approach Cochrane Handbook (2011) defines systematic review as a review of a clearly formulated question that uses systematic and explicit methods to identify, select and critically appraise relevant research and to collect and analyze data from the studies that are included in the review. In this paper, search results of systematic reviews on a clinical topic between two major databases – PubMed and Cochrane Library – are compared. Findings Searching within PubMed for key terms in the titles and abstracts of articles is important to include in any systematic review, in addition to searching Medical Subject Heading terms. After applying filters, PubMed retrieved 130 systematic reviews that matched the criteria. In Cochrane Library, the searches were performed on the chosen topic using Boolean and phrase searching: text field searches resulted in 251 reviews. The search was further narrowed by subject, which yielded 20 reviews. It is strongly recommended to use multiple health-care specialty databases, check for duplicate reviews in the results and not limit results to English-only publications. Practical implications This paper can be used to introduce new researchers and/or students to methods for conducting systematic reviews using two or more databases on a chosen topic. Originality/value This paper fills a gap in the literature regarding comparative efficiencies of major databases for systematic review results on clinically related topics.

Purpose The purpose of this paper is to present an efficient and scalable Arabic semantic search engine based on a domain-specific ontological graph for Colleges of Applied Science, Sultanate of Oman (CASOnto). It also supports the factorial question answering and uses two types of searching: the keyword-based search and the semantics-based search in both languages Arabic and English. This engine is built on variety of technologies such as resource description framework data and ontological graph. Furthermore, two experimental results are conducted; the first is a comparison among entity-search and the classical-search in the system itself. The second compares the CASOnto with well-known semantic search engines such as Kngine, Wolfram Alpha and Google to measure their performance and efficiency. Design/methodology/approach The design and implementation of the system comprises the following phases, namely, designing inference, storing, indexing, searching, query processing and the user’s friendly interface, where it is designed based on a specific domain of the IBRI CAS (College of Applied Science) to highlight the academic and nonacademic departments. Furthermore, it is ontological inferred data stored in the tuple data base (TDB) and MySQL to handle the keyword-based search as well as entity-based search. The indexing and searching processes are built based on the Lucene for the keyword search, while TDB is used for the entity search. Query processing is a very important component in the search engines that helps to improve the user’s search results and make the system efficient and scalable. CASOnto handles the Arabic issues such as spelling correction, query completion, stop words’ removal and diacritics removal. It also supports the analysis of the factorial question answering. Findings In this paper, an efficient and scalable Arabic semantic search engine is proposed. The results show that the semantic search that built on the SPARQL is better than the classical search in both simple and complex queries. Clearly, the accuracy of semantic search equals to 100 per cent in both types of queries. On the other hand, the comparison of CASOnto with the Wolfram Alpha, Kngine and Google refers to better results by CASOnto. Consequently, it seems that our proposed engine retrieved better and efficient results than other engines. Thus, it is built according to the ontological domain-specific, highly scalable performance and handles the complex queries well by understanding the context behind the query. Research limitations/implications The proposed engine is built on a specific domain (CAS Ibri – Oman), and in the future vision, it will highlight the nonfactorial question answering and expand the domain of CASOnto to involve more integrated different domains. Originality/value The main contribution of this paper is to build an efficient and scalable Arabic semantic search engine. Because of the widespread use of search engines, a new dimension of challenge is created to keep up with the evolution of the semantic Web. Whereas, catering to the needs of users has become a matter of paramount importance in the light of artificial intelligence and technological development to access the accurate and the efficient information in less possible time. However, the research challenges still in its infancy due to lack of research engine that supports the Arabic language. It could be traced back to the complexity of the Arabic language morphological and grammar rules.

SEO stands for "search engine optimization." It is the process of getting traffic from the free, organic, editorial or natural search results on search engines. The earlier (or higher ranked on the search results page), and more frequently a site appears in the search results list, the more visitors it will receive from the search engine's users, and these visitors can be converted into customers. SEO may target different kinds of search, including image search, local search, video search, academic search, news search and industry-specific vertical search engines. As an Internet marketing strategy, SEO considers how search engines work, what people search for, the actual search terms or keywords typed into search engines and which search engines are preferred by their targeted audience. Optimizing a website may involve editing its content, HTML and associated coding to both increase its relevance to specific keywords and to remove barriers to the indexing activities of search engines. Promoting a site to increase the number of backlinks, or inbound links, is another SEO tactic. As of May 2015, mobile search has finally surpassed desktop search, Google is developing and pushing mobile search as the future in all of its products and many brands are beginning to take a different approach on their internet strategies.

In this research study, the authors investigate the association between external searching, which is searching on a web search engine, and internal searching, which is searching on a website. They classify 295,571 external – internal searches where each search is composed of a search engine query that is submitted to a web search engine and then one or more subsequent queries submitted to a commercial website by the same user. The authors examine 891,453 queries from all searches, of which 295,571 were external search queries and 595,882 were internal search queries. They algorithmically classify all queries into states, and then clustered the searching episodes into major searching configurations and identify the most commonly occurring search patterns for both external, internal, and external-to-internal searching episodes. The research implications of this study are that external sessions and internal sessions must be considered as part of a continuous search episode and that online businesses can leverage external search information to more effectively target potential consumers.

Recent years have witnessed the emergence of new types of semantic search engines which attempt to overcome the defects of the traditional search engines by providing different search patterns. A big question here is that in order to achieve the semantic search engines SSEs, what types of search patterns should SSEs support? To help seek one of the many possible answers, in this paper we start with classifying and comparing current search engines, particularly from the perspective of search patterns which consist of index structure, user profiles, and interaction mechanism. We then present a novel search pattern named ExNa by defining its model and basic operations in detail. To validate the ExNa search pattern, we develop a prototype search engine named KNOWLE, and the experimental results show that KNOWLE equipped with ExNa can improve both the efficiency of the entire system when compared with search engines of other search patterns. Copyright © 2016 John Wiley & Sons, Ltd.

A booming cloud service has made it possible to consider a third-party server as private storage. When storage is used as a document archive, it is useful to provide functionality that allows users to find documents containing a specified string as a substring. The current method for providing such functionality while keeping the contents of the documents secret, either has linear complexity in the total size of the documents or needs to extract keywords when the documents are stored. Therefore, no method exists for efficiently searching for an arbitrary string that has not been extracted as a keyword. We propose the first encryption scheme that enables efficient searching for an arbitrary string. Although our scheme leaks some information for the sake of efficiency, all information leaked is derived from the precise leakage profile we have defined. ©i¾?2016 The Authors Security and Communication Networks Published by John Wiley & Sons Ltd

The study aims to find out whether important search techniques could be employed to find records in the ten online public access catalogs (OPACs) tested. The following search tests are conducted: (1) test to determine the fields used as access point to records, and the number of fields searched when a two-word syntax was used; (2) search using one to three charaters, one word, and English stopwords; (3) search using boolean AND, OR or NOT and their combinations; (4) phrase search; (5) truncation search; and (6) search using filters (e.g., language or year of publication) provided by the OPACs. Results of the tests were: (1) access to records was available through title, author, subject, series, and note in eight systems; (2) a two-word syntax was searched only in one field in all systems; (3) English words and syntaxes of one to three characters were searched in eight systems; (4) boolean operators AND, OR and NOT and their combinations could be used only in one system; (5) search employing filters could be performed in three systems. It was concluded that nine of the ten OPACs could not provide optimal access to records. The OPAC is the only access to library collections for those who search through the Internet; therefore, complete search tests of its search facilities are required before a system is released. Further research is needed to find efficient search test methods that could be used by the developers and librarians to prove that a search facility functions.

Search moves for finding novels in five search tasks and two catalogs were analyzed. Search tasks reflected the following search tactics: known-author search, topical search, open-ended browsing, search by analogy, and searching without a query. The most used search moves in both catalogs across all tasks were querying, search results inspection, and book page examination. In a traditional catalog, more effort was needed in the form of queries, search moves, and opened book pages to gain equivalent average book scores when compared with an enriched catalog. In a traditional catalog, a typical search strategy for interesting titles seemed to involve issuing queries and considering suitable entry terms carefully, and devoting more attention to search results instead of book pages. In an enriched catalog, a common approach involved time devoted to exploring the catalog's enriched front page and multiple entry points together with attention to the enriched results list.

Nowadays, due to the explosive growth of web content and usage, users deal with their complex search tasks by web search engines. However, conventional search engines consider a search query corresponding only to a simple search task. In order to accomplish a complex search task, which consists of multiple subtask search goals, users usually have to issue a series of queries. For example, the complex search task “travel to Dubai” may involve several subtask search goals, including reserving hotel room, surveying Dubai landmarks, booking flights, and so forth. Therefore, a user can efficiently accomplish his or her complex search task if search engines can predict the complex search task with a variety of subtask search goals. In this work, we propose a complex search task model (CSTM) to deal with this problem. The CSTM first groups queries into complex search task clusters, and then generates subtask search goals from each complex search task cluster. To raise the performance of CSTM, we exploit four web resources including community question answering, query logs, search engine result pages, and clicked pages. Experimental results show that our CSTM is effective in identifying the comprehensive subtask search goals of a complex search task.

Lexical embeddings can serve as useful representations for words for a variety of NLP tasks, but learning embeddings for phrases can be challenging. While separate embeddings are learned for each word, this is infeasible for every phrase. We construct phrase embeddings by learning how to compose word embeddings using features that capture phrase structure and context. We propose efficient unsupervised and task-specific learning objectives that scale our model to large datasets. We demonstrate improvements on both language modeling and several phrase semantic similarity tasks with various phrase lengths. We make the implementation of our model and the datasets available for general use.

The sensitivity of information is dependent on the context of application and user preference. Protecting sensitive data in the cloud era requires identifying them in the first place. It typically needs intensive manual efforts. More importantly, users may specify sensitive information only through an implicit manner. Existing research efforts on identifying sensitive data from its descriptive texts focus on keyword/phrase searching. These approaches can have high false positives/negatives as they do not consider the semantics of the descriptions. In this paper, we propose S3, an automated approach to identify sensitive data based on users’ implicit specifications. Our approach considers semantic, syntactic and lexical information comprehensively, aiming to identify sensitive data by the semantics of its descriptive texts. We introduce the notion concept space to represent the user’s notion of privacy, by which our approach can support flexible user requirements in defining sensitive data. Our approach is able to learn users’ preferences from readable concepts initially provided by users, and automatically identify related sensitive data. We evaluate our approach on over 18,000 top popular applications from Google Play Store. S3 achieves an average precision of 89.2%, and average recall 95.8% in identifying sensitive data.

Search in meta-data is challenging due to the sparsity of the available textual information. To alleviate the sparsity problem, the paper in hand evolves from the existing document expansion paradigm and proposes pseudo-descriptions as a new paradigm. Instead of encoding paradigmatic term relations implicitly in an expansion vector, we generate an explicit cohesive text field for meta-data records that describes the entity associated with the record. In contrast to document expansions, pseudo-descriptions allow to reveal why a certain document is considered relevant although the original meta-data does not contain the query terms. Moreover, they are easier to operationalize and facilitate the use of sophisticated retrieval features such as phrase search and query term proximity. To generate pseudo-descriptions, we propose a relevance dependent strategy that depends on the search engine result pages obtained from issuing the meta-data as a search query to a designated reference search engine. To demonstrate the validity of the pseudo-description paradigm, we experiment with different TREC collections where we withhold the content information to simulate a meta-data retrieval scenario. Though retrieval with full content information remains superior, our approach achieves retrieval performance improvements en par with document expansion.

With the widespread usage of Cloud Computing model in business, organizations prefer to outsource their data to the cloud. In order to preserve confidentiality of data, the data is encrypted before being outsourced. In order to use this encrypted data in decision making, symmetric searchable encryption is employed that enables users to fetch data matching the keywords over encrypted data without decrypting the data. Many existing schemes search for a single keyword without an index, leading to search time linear to the size of data. And moreover, support for multiple keyword searches or phrase search is also a trending challenge in the field of searchable encryption. In this paper, we extend the practical techniques for searches on encrypted data proposed by Song, et.al, by improving the search efficiency with help of an encrypted and dynamic index. We use a very efficient structure to enable dynamic updates to the index in event of changes to the encrypted dataset. Our proposed scheme is the first one to use such a simple yet update-efficient structure for index. Multi-phrase search being implemented helps in obtaining most representative documents for searches, especially in mobile devices. We also implemented and analyzed efficiency of Song et al scheme, which is one of the pioneering works in field of searchable encryption. Security analysis and extensive experiments were conducted to test the security guarantees and efficiency of the proposed extensions.

The study set out to examine the researchers’ perception towards the use JSTOR database and to explore the reasons not to use JSTOR database. The present study carried out using simple random sampling technique. The major purpose of researchers to use JSTOR was research work and article title was the most common searching techniques in simple search category. In advanced search technique field search was most common search technique followed by phrase search. Almost, all respondents found satisfied with the service of JSTOR and none was found dissatisfied/partially satisfied. On the basis of findings, it is recommended that the use of JSTOR database can be maximised, if researchers have frequent literacy programs.