Academic Search System Research Articles

Diagnostic Accuracy of AI for Opportunistic Screening of Abdominal Aortic Aneurysm in CT: A Systematic Review and Narrative Synthesis.

In this review, we focused on the applicability of artificial intelligence (AI) for opportunistic abdominal aortic aneurysm (AAA) detection in computed tomography (CT). We used the academic search system PubMed as the primary source for the literature search and Google Scholar as a supplementary source of evidence. We searched through 2 February 2022. All studies on automated AAA detection or segmentation in noncontrast abdominal CT were included. For bias assessment, we developed and used an adapted version of the QUADAS-2 checklist. We included eight studies with 355 cases, of which 273 (77%) contained AAA. The highest risk of bias and level of applicability concerns were observed for the "patient selection" domain, due to the 100% pathology rate in the majority (75%) of the studies. The mean sensitivity value was 95% (95% CI 100-87%), the mean specificity value was 96.6% (95% CI 100-75.7%), and the mean accuracy value was 95.2% (95% CI 100-54.5%). Half of the included studies performed diagnostic accuracy estimation, with only one study having data on all diagnostic accuracy metrics. Therefore, we conducted a narrative synthesis. Our findings indicate high study heterogeneity, requiring further research with balanced noncontrast CT datasets and adherence to reporting standards in order to validate the high sensitivity value obtained.

Design and implementation of an academic expert system through big data analysis

Most researchers establish research directions in their study of new fields by providing expert advice or publishing expert papers. The existing academic search services display papers by field but do not provide experts by field. Therefore, researchers are left to judge experts in each field by analyzing the papers for themselves. In this paper, we design and implement an expert search system based on papers that have been published in the academic societies. The academic expert search system is based on a big data processing system to handle a large amount of data in academic fields. It calculates an expert score using quality and influence factors. The quality factor is calculated based on the citations, impact factor, and recentness of a paper. The influence factor is measured by the sparsity of a field and the degree of contributiveness of an author. The proposed system provides various services such as expert searches, keyword searches, the hot topics, expert relationships, and academic society statistics. By finding experts in a specific field, our system can support researchers’ research activities.

CONNA: Addressing Name Disambiguation on the Fly

Name disambiguation is a key and also a very tough problem in many online systems such as social search and academic search. Despite considerable research, a critical issue that has not been systematically studied is <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">disambiguation on the fly</i> — to complete the disambiguation in the real-time. This is very challenging, as the disambiguation algorithm must be accurate, efficient, and error tolerance. In this paper, we propose a novel framework — CONNA — to train a matching component and a decision component jointly via reinforcement learning. The matching component is responsible for finding the top matched candidate for the given paper, and the decision component is responsible for deciding on assigning the top matched person or creating a new person. The two components are intertwined and can be bootstrapped via jointly training. Empirically, we evaluate CONNA on two name disambiguation datasets. Experimental results show that the proposed framework can achieve a 1.21-19.84 percent improvement on F1-score using joint training of the matching and the decision components. The proposed CONNA has been successfully deployed on AMiner — a large online academic search system.

Query Optimization of Data Based on Window Function and Distributed Cluster in Visual Academic Search System

Query Optimization of Data Based on Window Function and Distributed Cluster in Visual Academic Search System

AMiner: Search and Mining of Academic Social Networks

AMiner is a novel online academic search and mining system, and it aims to provide a systematic modeling approach to help researchers and scientists gain a deeper understanding of the large and heterogeneous networks formed by authors, papers, conferences, journals and organizations. The system is subsequently able to extract researchers’ profiles automatically from the Web and integrates them with published papers by a way of a process that first performs name disambiguation. Then a generative probabilistic model is devised to simultaneously model the different entities while providing a topic-level expertise search. In addition, AMiner offers a set of researcher-centered functions, including social influence analysis, relationship mining, collaboration recommendation, similarity analysis, and community evolution. The system has been in operation since 2006 and has been accessed from more than 8 million independent IP addresses residing in more than 200 countries and regions.

To be the Prince to wake up Sleeping Beauty: the rediscovery of the delayed recognition studies

In science, sleeping papers, previously known as “Sleeping Beauties”, refer to scientific papers that are recognized by the scientific community after a long hibernation period following their publication. Many factors may contribute to their delayed yet exceptional popularity, such as the introduction of new technologies or ideas that are beyond the capabilities at the time of publication. The recognition of a sleeping paper, often through a paper that cites the sleeping paper and has a profound impact on the research area, is important to the scientific community. Here, we proposed a method to identify the paper that rediscovers a sleeping paper, known as a rediscovering paper, based on the citation network of the sleeping paper. Based on the 15 rediscovering papers obtained from the top sleeping papers in science, we introduced 5 feature indices of the leading authors of these rediscovering papers (rediscovering authors) defined by an academic search system AMiner ( https://cn.aminer.org/ ). The 5 feature indices depict academic achievements of researchers from various aspects: Publication, Citation, Longevity, H-index and Sociability. The rediscovering authors lead to most general scientific authors in the 5 feature indices. Our results reveal common features of potential rediscovering authors in the scientific community who may play significant roles in the propagation of citation networks.

The Odysci Academic Search System

Web-based academic search systems have enabled researchers to find and access scientific publications worldwide in a way that was not possible ten years ago. Nevertheless, the quality of such search systems can vary considerably in coverage, ranking quality, entity resolution and indexing time. This paper describes the Odysci Academic Search System including all steps necessary from acquiring a document to making it available for user search. We describe in detail the steps for data import, data de-duplication, data indexing, search process, ranking of results, parallel processing and finally user experience, including many practical aspects commonly overlooked in publications, but important to make the system work well and provide a good user experience.

Emerging of Academic Information Search System with Ontology-based Approach

This paper describes the comparison of ontology development tools for development of academic information search system that assists inexperienced research students at a local university in Malaysia to search for academic resources in the local language context (Bahasa Malaysia). The cohort of inexperienced research students faces two main problems when using current system comprises of keyword search. Firstly the language barrier-limiting students’ capabilities to conduct keyword search in foreign language (such as English). Secondly limited research experience in querying often results in obtaining irrelevant search results. The proposed semantic search system aims to apply ontology-based search to overcome the above two problems. The paper presents the first phase of system development; ontology design and ontology development tool.

A Study on Developing and Refining a Large Citation Service System

Today, citation index information is used as an outcome scale of spreading technology and encouraging research. Article citation information is an important factor to determine the authority of the relevant author. Google Scholar uses the article citation information to organize academic article search results with a rank algorithm. For an accurate analysis of such important citation index information, large amounts of bibliographic data are required. Therefore, this study aims to build a fast and efficient system for large amounts of bibliographic data, and to design and develop a system for quickly analyzing cited information for that data. This study also aims to use and analyze citation data to be a basic element for providing various advanced services to the academic article search system.

전거데이터 기반 학술논문 검색시스템의 구축 및 평가에 관한 연구

This study aims to develop and evaluate the authority data-based academic retrieval system that may provide search tools and additional information for search of academic papers with users. To evaluate the usability of authority data-based academic paper search system, this work studied 60 subjects who had experienced the academic paper service to measure their satisfaction on both search efficiency and system usability and additionally analyzed whether the difference between search efficiency and system availability is significant in the experimental search system among user groups. As a result, the authority data-based academic paper search system was analyzed to show a higher search efficiency and user satisfaction than the ordinary system.

Topic level expertise search over heterogeneous networks

In this paper, we present a topic level expertise search framework for heterogeneous networks. Different from the traditional Web search engines that perform retrieval and ranking at document level (or at object level), we investigate the problem of expertise search at topic level over heterogeneous networks. In particular, we study this problem in an academic search and mining system, which extracts and integrates the academic data from the distributed Web. We present a unified topic model to simultaneously model topical aspects of different objects in the academic network. Based on the learned topic models, we investigate the expertise search problem from three dimensions: ranking, citation tracing analysis, and topical graph search. Specifically, we propose a topic level random walk method for ranking the different objects. In citation tracing analysis, we aim to uncover how a piece of work influences its follow-up work. Finally, we have developed a topical graph search function, based on the topic modeling and citation tracing analysis. Experimental results show that various expertise search and mining tasks can indeed benefit from the proposed topic level analysis approach.

LibraryFind: System design and usability testing of academic metasearch system

AbstractUsing off‐the‐shelf search technology provides a single point of access into library resources, but we found that such commercial systems are not entirely satisfactory for the academic library setting. In response to this, Oregon State University (OSU) Libraries designed and deployed LibraryFind, a metasearch system. We conducted a usability experiment comparing LibraryFind, the OSU Libraries Web site, and Google Scholar. Each participant used all three search systems in a controlled setting, and we recorded their behavior to determine the effectiveness and efficiency of each search system. In this article, we focus on understanding what factors are important to undergraduates in choosing their primary academic search system for class assignments. Based on a qualitative and quantitative analysis of the results, we found that mimicking commercial Web search engines is an important factor to attract undergradu‐ates; however, when undergraduates use these kinds of search engines, they expect similar performance to Web search engines, including factors such as relevance, speed, and the availability of a spell checker. They also expected to be able to find out what kinds of content and materials are available in a system. Participants' prior experience using academic search systems also affected their expectations of a new system.