Information Retrieval Methods Research Articles

The Glossaryfication Web Service – an automated glossary creation tool to support One Health communication

Efficient communication and collaboration across sectors is an important precondition for true One Health Surveillance (OHS) activities. Despite the overall willingness to embrace the One Health paradigm, it is still challenging to accomplish this in day-to-day practice due to the differences in terminology and interpretation of sector-specific terms. In this sense, simple interventions like the inclusion of integrative glossaries in OHS documents (e.g. reports, research papers and guidelines) would help to reduce misunderstandings and could significantly improve the written communication in OHS. Here, we present the Glossaryfication Web Service that generates a document-specific glossary for any text file provided by the user. The web service automatically adds the available definitions with their corresponding references for the words in the document that match with terms in the user-selected glossaries. The Glossaryfication Web Service was developed to provide added value to the OHEJP Glossary that was developed within the OHEJP project ORION. The OHEJP Glossary improves the communication and collaboration among OH sectors by providing an online resource that lists relevant OH terms and sector-specific definitions. The Glossaryfication Web Service supports the practical use of the curated OHEJP Glossary and can also source information from other glossaries relevant for OH professionals (currently supporting the online CDC, WHO and EFSA glossaries). The Glossaryfication Web Service was created using the open-source software KNIME and the KNIME Text Processing extension (https://www.knime.com/knime-text-processing). The Glossaryfication KNIME workflow is deployed on BfR’s KNIME Server infrastructure providing an easy-to-use web interface where the users can upload their documents (any text-type file e.g. PDF, Word, Excel) and select the desired glossary to compare with. The Glossaryfication KNIME workflow reads in the document provided via the web interface and applies natural language processing (e.g. text cleaning, stemming), transforming (bag-of-words generation) and information retrieval methods to identify the matching terms in the selected glossaries. The Glossaryfication Web Service generates as an output a table containing all the terms that match with the selected glossaries. It also provides the available definitions, corresponding references and additional meta-information, e.g. the term frequency, i.e., how often each term appears in the given text, and the sectoral classification (only for the OHEJP Glossary terms). Furthermore, the workflow generates a tag cloud where the terms are categorized as: (i) exact match when the term in the text matches exactly with the entry of this term in the glossary; (ii) inexact match when the term appears in the text slightly modified (e.g. plural forms or suffixes) and (iii) non-matching that corresponds to all the other words appearing in the text that do not match with any glossary term. Through the user interface, the users can then choose if they want to download the whole list of terms, select only the exact/inexact matching terms, or just choose those terms and definitions that match with the meaning intended for this term in the user-provided document. The resulting table of terms can be downloaded as an Excel file and added to the user’s document as a document-specific glossary. The Glossaryfication Web Service provides an easy-to-adopt solution to enrich documents and reports with more comprehensive and unambiguous glossaries. Furthermore, it improves the referentiality of terms and definitions from different OH sectors. An additional feature provided by the Glossaryfication Web Service is the possibility of extending its use to other glossaries from other national or international institutions allowing the user to customize this glossary creation service.

Heritage connector: A machine learning framework for building linked open data from museum collections

AbstractAs with almost all data, museum collection catalogues are largely unstructured, variable in consistency and overwhelmingly composed of thin records. The form of these catalogues means that the potential for new forms of research, access and scholarly enquiry that range across multiple collections and related datasets remains dormant. In the project Heritage Connector: Transforming text into data to extract meaning and make connections, we are applying a battery of digital techniques to connect similar, identical and related objects within and across collections and other publications. In this article, we describe a framework to create a Linked Open Data knowledge graph from digital museum catalogues, perform record linkage to Wikidata, and add new entities to this graph from textual catalogue record descriptions (information retrieval). We focus on the use of machine learning to create these links at scale with a small amount of labelled data, and models which are small enough to run inference on datasets the size of museum collections on a mid‐range laptop or a small cloud virtual machine. Our method for record linkage against Wikidata achieves 85%+ precision with the Science Museum Group (SMG) collection, and our method for information retrieval is shown to improve NER performance compared with pretrained models on the SMG collection with no labelled training data. We publish open‐source software providing tools to perform these tasks.

A comparative analysis of system features used in the TREC-COVID information retrieval challenge

The COVID-19 pandemic has resulted in a rapidly growing quantity of scientific publications from journal articles, preprints, and other sources. The TREC-COVID Challenge was created to evaluate information retrieval (IR) methods and systems for this quickly expanding corpus. Using the COVID-19 Open Research Dataset (CORD-19), several dozen research teams participated in over 5 rounds of the TREC-COVID Challenge. While previous work has compared IR techniques used on other test collections, there are no studies that have analyzed the methods used by participants in the TREC-COVID Challenge. We manually reviewed team run reports from Rounds 2 and 5, extracted features from the documented methodologies, and used a univariate and multivariate regression-based analysis to identify features associated with higher retrieval performance. We observed that fine-tuning datasets with relevance judgments, MS-MARCO, and CORD-19 document vectors was associated with improved performance in Round 2 but not in Round 5. Though the relatively decreased heterogeneity of runs in Round 5 may explain the lack of significance in that round, fine-tuning has been found to improve search performance in previous challenge evaluations by improving a system’s ability to map relevant queries and phrases to documents. Furthermore, term expansion was associated with improvement in system performance, and the use of the narrative field in the TREC-COVID topics was associated with decreased system performance in both rounds. These findings emphasize the need for clear queries in search. While our study has some limitations in its generalizability and scope of techniques analyzed, we identified some IR techniques that may be useful in building search systems for COVID-19 using the TREC-COVID test collections.

Загальні принципи обґрунтування ефективності інформаційно-вимірювальних технологій в процесі експлуатації об'єктів нерухомості

Aim of this work – to set forth general principles of ground of efficiency of informatively-measuring technologies in the process of exploitation of objects of the real estate. Methodology. The method of substantiating the effectiveness of the restoration of the technical condition of real estate in the process of their operation is based on the use of information and measurement technologies to establish the technical condition of the object, which is a combination of information retrieval methods, instrumental measurement, analysis and processing of such information, technological and technical solutions to ensure the operational suitability of the facility in conditions of uncertainty. The method involves the formation of a repair strategy using information and measurement technologies based on the development of the life cycle of repairs and algorithms for selecting options for instrumental monitoring systems, choosing the procedure for studying the duration of measurement work, justification of labor and information technology costs and comparative assessment of total costs of real estate object. Results. It is established that measuring works are an integral part of the process of operation and repair work in its composition, which are carried out according to a single schedule of operation of the building and are part of repair work, and the frequency and volume of measuring, including geodetic work correlate with. To plan the volume of measuring work, their duration and labor costs are determined on the basis of existing time norms, and in their absence – by methods of technical rationing. Substantiation of the effectiveness of information and measurement technologies in the operation of real estate is performed by comparative assessment of total costs during the operational stage of its life cycle, taking into account the costs of implementing information and measurement technologies, that is the set of processes for obtaining information by instrumental measurement and processing ‘effective conclusion on the technical condition of the property. The scientific novelty of the obtained results is in the development of general principles for substantiating the effectiveness of information and measurement technologies in the operation of real estate. The practical significance lies in the improvement of standard solutions for the choice of composition and content of information and measurement technologies, taking into account their impact on the duration and labor costs of all types of repairs and reconstruction.

Intellectual systems in the management of medical technologies and quality of life

The article is devoted to the development and use of intelligent systems in the management of medical technological processes and health-related quality of life (HRQOL). The relevance of the work is due to the need for effective use of intelligent systems in healthcare. The purpose of this work is to study the possibilities and prospects of using information technologies and artificial intelligence systems in clinical medicine to improve the efficiency of providing medical care to the population. Information retrieval method; theoretical analysis of legislative and regulatory documents, literary sources, Internet resources, research results; spectral-dynamic and mathematical analysis of the current state and assessment of the quality of life of an individual using the artificial intelligence system "CME". The paper analyzes the development trends of information technologies and artificial intelligence systems, as well as the features of their use in medical technological processes. As an example, the technological capabilities of the intelligent system Complex Medical Expert are briefly described.

Класифікація джерел мікрозабруднювачів як складових забруднення поверхневих дорожніх стоків

Introduction. The current state and the development of road construction and infrastructure is the cause of increasing the environmental pollution, especially in close proximity to the roads.Problem Statement. Surface runoff from the roads is a significant volume of polluted water which most often enters water bodies and the surrounding area without treatment, which is contrary to environmental requirements. Not only accidental leaks of oil and chloride residues present in runoff, but also a significant volume of suspended solids, which are often settled and accumulated and are dangerous for the environment and public health. In order to predict measures or prevent such pollution and accumulation, it is necessary to determine the sources of suspended solids, qualitative characteristics, physical and chemical properties of pollutants.Purpose. The purpose of the work is to determine of suspended solids pollutants (micropollutants), their components, to develop a classification of sources and their formation.Materials and methods. Analytical, information retrieval methods were used for this study.Results. As a result of the work, classification, which presents potentially possible types of micropollutants which are the components of suspended solids found in road surface runoff, was developed.Conclusions. The developed classification of micropollutants sources is the initial stage to study such kind of pollution, migration in the environment, the impacts of accumulation and decomposition products, as well as to develop measures for reducing the formation and dissemination of micropollutants in the environment.Keywords: road pavement, micropollutants, composition, particles.

Basic principles of information security monitoring in physical channels of optical access networks

The article presents theoretical information about the methods of information retrieval from optical fibers. The simplest analyzed method is the bending of the optical fiber when violating the conditions of total internal reflection. The functional dependence of the losses and the level of radiated power on the parameters and bending geometry of the optical fiber are given. In this case, the introduced losses are monitored in the monitoring system for detecting fiber bends. The structure, principles of operation, and trends in the development of monitoring systems are presented. The main technical characteristics of optical radiation detection devices are demonstrated.

The hybridised indexing method for research-based information retrieval

An information retrieval system (IRS) is used to retrieve documents based on an information need. The IRS makes relevance judgements by attempting to match a query to a document. As IRS capabilities are indexing design dependent, the hybrid indexing method (IRS-H) is introduced. The objectives of this article are to examine IRS-H (as an alternative indexing method that performs exact phrase matching) and IRS-I, regarding retrieval usefulness, identification of relevant documents, and the quality of rejecting irrelevant documents by conducting three experiments and by analysing the related data. Three experiments took place where a collection of 100 research documents and 75 queries were presented to: (1) five participants answering a questionnaire, (2) IRS-I to generate data and (3) IRS-H to generate data. The data generated during the experiments were statistically analysed using the performance measurements of Precision, Recall and Specificity, and one-tailed Student’s t-tests. The results reveal that IRS-H (1) increased the retrieval of relevant documents, (2) reduced incorrect identification of relevant documents and (3) increased the quality of rejecting irrelevant documents. The research found that the hybrid indexing method, using a small closed document collection of a hundred documents, produced the required outputs and that it may be used as an alternative IRS indexing method.

Mining Likely Analogical APIs Across Third-Party Libraries via Large-Scale Unsupervised API Semantics Embedding

Establishing API mappings between third-party libraries is a prerequisite step for library migration tasks. Manually establishing API mappings is tedious due to the large number of APIs to be examined. Having an automatic technique to create a database of likely API mappings can significantly ease the task. Unfortunately, existing techniques either adopt supervised learning mechanism that requires already-ported or functionality similar applications across major programming languages or platforms, which are difficult to come by for an arbitrary pair of third-party libraries, or cannot deal with lexical gap in the API descriptions of different libraries. To overcome these limitations, we present an unsupervised deep learning based approach to embed both API usage semantics and API description (name and document) semantics into vector space for inferring likely analogical API mappings between libraries. Based on deep learning models trained using tens of millions of API call sequences, method names and comments of 2.8 millions of methods from 135,127 GitHub projects, our approach significantly outperforms other deep learning or traditional information retrieval (IR) methods for inferring likely analogical APIs. We implement a proof-of-concept website ( https://similarapi.appspot.com ) which can recommend analogical APIs for 583,501 APIs of 111 pairs of analogical Java libraries with diverse functionalities. This scale of third-party analogical-API database has never been achieved before.

Information Iterative Retrieval of Internet of Things Communication Terminal Based on Symmetric Algorithm

Through improved similarity analysis algorithm and hybrid local sensitivity hashing method, information is summarized and retrieved by information retrieval method based on EPC network architecture. This method does not consider the symmetry of information, and its retrieval effect is poor, security is low, and resource occupancy rate is high. Therefore, the information iterative retrieval of IoT communication terminal based on symmetric algorithm is proposed to solve these problems. The symmetric private information retrieval protocol based on secret sharing is used to obtain the information set with retrieval, and the Paillier homomorphic information retrieval method is used to obtain the original document of the information communication terminal information. Then, the information detection model based on the iterative algorithm is used to obtain information of the IoT communication terminal. The experimental results show that the retrieval accuracy of the proposed method is higher than 96% when the document set is less than 1000 KB, and the retrieval accuracy is 94.47% when the document set is 10 TB. The CPU usage and memory usage of this method are 8.65 and 3.24%, respectively. Experiments show that the proposed method has extremely high retrieval accuracy and low energy consumption.

Can the application of certain music information retrieval methods contribute to the machine learning classification of electrocardiographic signals?

The electrocardiogram is traditionally used to diagnose a large number of heart pathologies. Research to improve the readability and classification of cardiac signals includes studies geared toward sonification of the electrocardiographic signal and others involving features related to music processing, such as Mel-frequency cepstral coefficients. In terms of music processing features, this study seeks to use music information retrieval (MIR) features as electrocardiographic signal descriptors. The study compares the discriminatory capability of the introduced features in relation to standard groups such as heart rate variability, wavelet transform, descriptive statistics, Mel coefficients and fractal analysis, evaluated using classification algorithms; the signals analyzed were extracted from public databases. The group of features extracted from wavelet transform and the MIR group showed a high level of discrimination; the best representation of the ECG signals in the study was achieved in most cases by the MIR features. Moreover, a correlation coefficient higher than 0.8 was found between a number of MIR and other feature groups, indicating a likely relationship between the electrocardiographic signals and MIR features. These results suggest the feasibility of representing the analyzed signals by music information retrieval descriptors, giving the potential to consider these electrocardiographic signals as analogues to musical signals.

Website information retrieval of web database based on symmetric encryption algorithm

With the diversification of information development, data security is becoming more and more important. Therefore, the research on website information retrieval of Web database based on symmetric encryption algorithm is studied. From the security performance of Website information retrieval of Web database, the improved chaotic map encryption algorithm is used to enhance the hyper chaos of the key. It is undeniable that digital signature can improve the security of Web database information authentication. Based on this; the Web information retrieval method based on mobile Agent is adopted. This method is a modular structure of multi-agents that is divided and cooperated by five types of Agents. Through the division and cooperation of the user interaction Agent, retrieval management Agent, evaluation screening Agent, preference acquisition Agent, and the network spider Agent, the user’s preference is selected. Finally, the effective retrieval of Web information is achieved. The experimental results show that this kind of retrieval method can comprehensively and effectively retrieve the information of various commodities in website of the logistics company, and the retrieval performance is better.

A novel index retrieval and query optimisation method for private information retrieval in location-based service application

Location-based service is a popular information and communications technology. Security, trust and privacy are the major concerns preventing the wide deployment of LBS. In this paper, we address privacy issues by employing computational private information retrieval schemes and highlight a few optimisation methods. We propose a novel index retrieval technique which helps the user to identify his grid ID and know the index value for the point-of-interest (POI) type of his interest, and an adaptive computation method (flip-optimisation) to reduce multiplication cost for PIR query used to retrieve the POI item at the specified index. The adaptive computation method proposed in this paper is generic and can be applied to any application which uses PIR protocol to access data privately. Our work empirically evaluated the proposed method by implementing the PIR prototype and found it suitable for a practical purpose.

A novel index retrieval and query optimisation method for private information retrieval in location-based service application

A novel index retrieval and query optimisation method for private information retrieval in location-based service application

Graph visualization of dark web hyperlinks and their feature analysis

Content regarding various illegal activities, such as weapon and drug trafficking, is shared on the dark web. Most of the illegal content is distributed on anonymous networks that cannot be directly accessed from the World Wide Web. A number of studies have been conducted on the network structure of the World Wide Web since its advent. Similar to the World Wide Web, the dark web is connected by hypertext transfer protocol (http); this makes it possible to use the methods developed for the web in the dark web. Many studies have investigated the dark web and its network structure. However, few studies have focused on the visualization of the dark web network structure, and there have been no studies investigating the temporal changes in the network structure. In this study, to understand the hypertext markup language (html) network structure of the dark web, we created and visualized a graph of the html hyperlink relations of the Tor network, which is popular on the dark web. We then compared the insights gained from graph centrality metrics with those gained from visualizations. The analyzed dataset comprised 25,270,157 pages of html text files crawled from the Tor network by breadth-first search from June 1, 2018, to January 30, 2021. Subsequently, we acquired half-yearly snapshots from the collected data and investigated the change in the dark web network over time using a time-series graph. Then, we derived the centrality metrics from the created graph data and confirmed the differences between the centrality metrics and visualizations. The results obtained in this study provided new insights into the dark web. First, we found that the dark web fluctuated significantly; the structure of the dark web network was more strongly interconnected. Second, most of the nodes that had increased in the past two years may have disappeared rapidly after May 2020. Third, analysis of each snapshot revealed that the proportion of highly volatile domains increased from 40% to 75% during the observation period. Fourth, after calculating the network centrality metrics from each snapshot and comparing the transition of hub nodes in chronological order, we observed that the importance of link-collection sites as the main information retrieval method used in the dark web decreased. Finally, we estimated the size of the dark web based on our observed dark web measurements using the mark-recapture method. To the best of our knowledge, this is the first study to use the mark-recapture method to estimate the size of the dark web network.

Peer review of searches for studies for health technology assessments, systematic reviews, and other evidence syntheses.

Peer review of searches is a process whereby both the search strategies and the search process description are reviewed, ideally using an evidence-based checklist. As the search strategy underpins any well-conducted evidence synthesis, its quality could affect the final result. Evidence shows, however, that search strategies are prone to error. There is increasing awareness and use of the PRESS Evidence-Based Checklist and peer review of search strategies, at the outset of evidence syntheses, prior to the searches being run, and this is now recommended by a number of evidence synthesis organizations. Searches for evidence syntheses should be peer reviewed by a suitably qualified and experienced librarian or information specialist after being designed, ideally, by another suitably qualified and experienced librarian or information specialist. Peer review of searches should take place at two important stages in the evidence synthesis process; at the outset of the project prior to the searches being run and at the prepublication stage. There is little empirical evidence, however, to support the effectiveness of peer review of searches. Further research is required to assess this. Those wishing to stay up to date with the latest developments in information retrieval, including peer review of searches, should consult the SuRe Info resource (http://www.sure-info.org), which seeks to help information specialists and others by providing easy access to the findings from current information retrieval methods research and thus support more research-based information retrieval practice.

Ontology-Based Analysis of Website Structure for Benchmarking in Retail Business

With the growing trend of digital transformation, electronic business has become a crucial part of business operations for many companies. Some of the most critical steps in successful transformation pertain not only to data and information acquisition and digitalization but also to adequate publishing and organization of these information resources. Companies share information through their websites, so the structuring of the web content becomes critical during the digital transformation of business. In this paper, the authors present a conceptual model of the managerial tool that is based on semantic analysis of web sites in order to obtain information about the structure of web sites in a particular domain. The resulting ontological model contains information about best practices in web information organization and can be a valuable resource for management when deciding on the organization of their own web content. The system is based on grounded theory and uses current information retrieval methods, natural language processing, semantic networks, and ontologies.

The Online Health Information Needs of Family Physicians: Systematic Review of Qualitative and Quantitative Studies.

BackgroundDigitalization and the increasing availability of online information have changed the way in which information is searched for and retrieved by the public and by health professionals. The technical developments in the last two decades have transformed the methods of information retrieval. Although systematic evidence exists on the general information needs of specialists, and in particular, family physicians (FPs), there have been no recent systematic reviews to specifically address the needs of FPs and any barriers that may exist to accessing online health information.ObjectiveThis review aims to provide an up-to-date perspective on the needs of FPs in searching, retrieving, and using online information.MethodsThis systematic review of qualitative and quantitative studies searched a multitude of databases spanning the years 2000 to 2020 (search date January 2020). Studies that analyzed the online information needs of FPs, any barriers to the accessibility of information, and their information-seeking behaviors were included. Two researchers independently scrutinized titles and abstracts, analyzing full-text papers for their eligibility, the studies therein, and the data obtained from them.ResultsThe initial search yielded 4541 studies for initial title and abstract screening. Of the 144 studies that were found to be eligible for full-text screening, 41 were finally included. A total of 20 themes were developed and summarized into 5 main categories: individual needs of FPs before the search; access needs, including factors that would facilitate or hinder information retrieval; quality needs of the information to hand; utilization needs of the information available; and implication needs for everyday practice.ConclusionsThis review suggests that searching, accessing, and using online information, as well as any pre-existing needs, barriers, or demands, should not be perceived as separate entities but rather be regarded as a sequential process. Apart from accessing information and evaluating its quality, FPs expressed concerns regarding the applicability of this information to their everyday practice and its subsequent relevance to patient care. Future online information resources should cater to the needs of the primary care setting and seek to address the way in which such resources may be adapted to these specific requirements.

Museum Mobile Guide Preferences of Different Visitor Personas

Personalising museum mobile guides is widely acknowledged as being important for enhancing the visitor experience. Due to the lack of information about an individual visitor and the relatively limited time of his or her visit, adapting the user interface based on a museum visitor's type is a promising approach to personalisation. This approach first requires a mechanism to identify the visitor type (‘persona’) and, second, knowledge of the preferences and needs of different types to apply personalisation. In this article, we report a face-to-face questionnaire study carried out with 105 visitors to Scitech, a science and technology visitor centre. The study aims to investigate the main facts required to identify a visitor persona and to explore the preferences of different visitor personas for particular mobile guide features. We limited our concern to the user interface features of the guide (e.g., whether it provides recommendations for related items to view) rather than what content and services the guide provides (e.g., what related items are recommended). We found that we can reliably identify the visitor persona using two multiple choice questions about visit motivation and perceived success criteria. In addition, we found that visitors have significant preferences for particular features such as presentation media, venue navigation tool, object suggestions, details level, accessing external links, exhibit information retrieval method and social interaction features such as voice communication, instant messaging, group games and challenges. Some features were found to be preferred differently by different personas such as the challenges feature, some were found to be preferred by personas differently to the overall preference such as in presentation media, and some were found to be preferred by some personas with no particular preference for others such as a venue navigation tool. Instant messaging was found to be significantly not preferred by all personas. The results provide a basis for personalisation of museum guides and services using a personas approach, which is a solution where data about individual users may be limited and where the individual configuration of a user interface may not be practical or warranted.

Musical features that aid sleep

Background: Music is thought to be beneficial as a sleep aid. However, little research has explicitly investigated the specific characteristics of music that aid sleep and some researchers assume that music described as generically sedative (slow, with low rhythmic activity) is necessarily conducive to sleep, without directly interrogating this assumption. This study aimed to ascertain the features of music that aid sleep. Method: As part of an online survey, 161 students reported the pieces of music they had used to aid sleep, successfully or unsuccessfully. The participants reported 167 pieces, some more often than others. Nine features of the pieces were analyzed using a combination of music information retrieval methods and aural analysis. Results: Of the pieces reported by participants, 78% were successful in aiding sleep. The features they had in common were that (a) their main frequency register was middle range frequencies; (b) their tempo was medium; (c) their articulation was legato; (d) they were in the major mode, and (e) lyrics were present. They differed from pieces that were unsuccessful in aiding sleep in that (a) their main frequency register was lower; (b) their articulation was legato, and (c) they excluded high rhythmic activity. Conclusion: Music that aids sleep is not necessarily sedative music, as defined in the literature, but some features of sedative music are associated with aiding sleep. In the present study, we identified the specific features of music that were reported to have been successful and unsuccessful in aiding sleep. The identification of these features has important implications for the selection of pieces of music used in research on sleep.