Keyword Classification Research Articles

Region Proposal Network Based Small-Footprint Keyword Spotting

We apply an anchor-based region proposal network (RPN) for end-to-end keyword spotting (KWS). RPNs have been widely used for object detection in image and video processing; here, it is used to jointly model keyword classification and localization. The method proposes several anchors as rough locations of the keyword in an utterance and jointly learns classification and transformation to the ground truth region for each positive anchor. Additionally, we extend the keyword/non-keyword binary classification to detect multiple keywords. We verify our proposed method on a hotword detection data set with two hotwords. At a false alarm rate of one per hour, our method achieved more than 15% relative reduction in false rejection of the two keywords over multiple recent baselines. In addition, our method predicts the location of the keyword with over 90% overlap, which can be important for many applications.

Trend Analysis of Theory-based Research Published in Asian Oncology Nursing

This study was conducted to identify and analyze the trends of theory-based research published in Asian Oncology Nursing (AON) from 2011 to 2017. Methods: We analyzed 22 theory-based studies taken from among the 232 studies published in AON for 7years. We analyzed the framework developed by researchers, descriptive summaries of theory, keyword classification of nursing meta-paradigms, the generic character of the studies, and the type of research designs. Results: A total of 23 theories were applied in the 22 (9.5%) studies. ‘Nursing’ was the most prevalent field of study (34.8%). In the field of study of theory, nursing theory was the largest with eight (34.8%), and in the method of theory utilization, all 22 studies were theoretical applications 3 middle-range theories were most prevalent. In the keyword classification using meta-paradigm in nursing, we found 83 key words. Among them, ‘Health’ related key words were most frequent (53), followed by ‘Cancer’ keywords (13). Conclusion: AON has a high ratio of theoretical study compared to other journals, but the ratio is still less than 10%, which is still weak. Therefore, it is believed that the attention to, utilization of, application, and institutional establishment of nursing theory should be necessary.

Critical analysis of classification system of partially edentulous spaces: A literature review

There are more than 65,000 possible combinations of standing teeth and edentulous spaces which are encountered in the human dental arch. The system of classifying the partially edentulous arch would be an enormous aid in providing a precise word picture of a particular dental arch that is to be discussed. However, there are numerous classification systems which create confusion. Hence, a broad search of published literature was performed electronically using the keyword classification, partially edentulous, and denture from January 1928 to December 2018. This article enlisted the various classification systems and its critical analysis.

An Intelligent Content Prefix Classification Approach for Quality of Service Optimization in Information-Centric Networking

This research proposes an intelligent classification framework for quality of service (QoS) performance improvement in information-centric networking (ICN). The proposal works towards keyword classification techniques to obtain the most valuable information via suitable content prefixes in ICN. In this study, we have achieved the intelligent function using Artificial Intelligence (AI) implementation. Particularly, to find the most suitable and promising intelligent approach for maintaining QoS matrices, we have evaluated various AI algorithms, including evolutionary algorithms (EA), swarm intelligence (SI), and machine learning (ML) by using the cost function to assess their classification performances. With the goal of enabling a complete ICN prefix classification solution, we also propose a hybrid implementation to optimize classification performances by integration of relevant AI algorithms. This hybrid mechanism searches for a final minimum structure to prevent the local optima from happening. By simulation, the evaluation results show that the proposal outperforms EA and ML in terms of network resource utilization and response delay for QoS performance optimization.

Identification of the CIRP expertise network based on public data

Identification of expertise of people, academic societies, companies, and institutions is an important professional and social activity for nurturing professional networks, forming project consortia, etc. Today, a lot of information about the skills and expertise is openly available from websites and public personal profiles on professional and scientific social networks. With computerised methods, such as web crawling, text mining, and keyword classification, it is possible to identify expertise networks from the raw unstructured data. The paper presents and analyses the extended expertise network of the CIRP scientific community and proposes a web-based tool for the structuring and management of the community’s expertise.

Vaporous Marketing: Uncovering Pervasive Electronic Cigarette Advertisements on Twitter.

BackgroundTwitter has become the “wild-west” of marketing and promotional strategies for advertisement agencies. Electronic cigarettes have been heavily marketed across Twitter feeds, offering discounts, “kid-friendly” flavors, algorithmically generated false testimonials, and free samples.MethodsAll electronic cigarette keyword related tweets from a 10% sample of Twitter spanning January 2012 through December 2014 (approximately 850,000 total tweets) were identified and categorized as Automated or Organic by combining a keyword classification and a machine trained Human Detection algorithm. A sentiment analysis using Hedonometrics was performed on Organic tweets to quantify the change in consumer sentiments over time. Commercialized tweets were topically categorized with key phrasal pattern matching.ResultsThe overwhelming majority (80%) of tweets were classified as automated or promotional in nature. The majority of these tweets were coded as commercialized (83.65% in 2013), up to 33% of which offered discounts or free samples and appeared on over a billion twitter feeds as impressions. The positivity of Organic (human) classified tweets has decreased over time (5.84 in 2013 to 5.77 in 2014) due to a relative increase in the negative words ‘ban’, ‘tobacco’, ‘doesn’t’, ‘drug’, ‘against’, ‘poison’, ‘tax’ and a relative decrease in the positive words like ‘haha’, ‘good’, ‘cool’. Automated tweets are more positive than organic (6.17 versus 5.84) due to a relative increase in the marketing words like ‘best’, ‘win’, ‘buy’, ‘sale’, ‘health’, ‘discount’ and a relative decrease in negative words like ‘bad’, ‘hate’, ‘stupid’, ‘don’t’.ConclusionsDue to the youth presence on Twitter and the clinical uncertainty of the long term health complications of electronic cigarette consumption, the protection of public health warrants scrutiny and potential regulation of social media marketing.

A review of brand research (1990-2010): classification, application and development trajectory

Facing rapidly changeable markets, a firm not only should make best use of its brands to strengthen recognition, attitude and purchase behaviour of consumers to enhance their purchase intention, but the firm also should build up and maintain B2B partnerships to achieve win-win situations. Unfortunately, the extant literature related to brands lacks systematic exploration and analyses. A comprehensive survey of extant brand literature was undertaken using keyword classification from 1990 to 2010. Based on 1,714 articles, we find that the number of publications on brands significantly increased since 2003, with steady growth in recent years and these selected articles are mostly published in 20 academic journals. In addition, there is no industry that has not implemented the concept of brand management and the majority of brand targeted countries come from Asia-Pacific emerging markets. This study ends with some discussions and conclusions.

Information systems frontiers: Keyword analysis and classification

The current study focuses on the disciplines of ISF researches. The various analyses have employed to address research discipline of Management Information Systems (MIS). Since ISF is a forum for not only academic researchers but also industrial experts, it is worth to shed light on distinctive research features of ISF. I created a keyword classification scheme, incorporating new research topics into Barki’s information systems keyword classification scheme. This research describes the disciplines of 318 papers in 14 issues published on ISF until 2012 by 796 scholars, examining word frequency and keyword co-occurrence.

Meta-analysis of scientometric research of knowledge management: discovering the identity of the discipline

Purpose – The purpose of this study is to conduct a meta-analysis of prior scientometric research of the knowledge management (KM) field. Design/methodology/approach – A total of 108 scientometric studies of the KM discipline were subjected to meta-analysis techniques. Findings – The overall volume of scientometric KM works has been growing, reaching up to ten publications per year by 2012, but their key findings are somewhat inconsistent. Most scientometric KM research is published in non-KM-centric journals. The KM discipline has deep historical roots. It suffers from a high degree of over-differentiation and is represented by dissimilar research streams. The top six most productive countries for KM research are the USA, the UK, Canada, Germany, Australia, and Spain. KM exhibits attributes of a healthy academic domain with no apparent anomalies and is progressing towards academic maturity. Practical implications – Scientometric KM researchers should use advanced empirical methods, become aware of prior scientometric research, rely on multiple databases, develop a KM keyword classification scheme, publish their research in KM-centric outlets, focus on rigorous research of the forums for KM publications, improve their cooperation, conduct a comprehensive study of individual and institutional productivity, and investigate interdisciplinary collaboration. KM-centric journals should encourage authors to employ under-represented empirical methods and conduct meta-analysis studies and should discourage conceptual publications, especially the development of new frameworks. To improve the impact of KM research on the state of practice, knowledge dissemination channels should be developed. Originality/value – This is the first documented attempt to conduct a meta-analysis of scientometric research of the KM discipline.

보육과정에서의 자유놀이 현황과 교사의 놀이인식

This study dealt with free play activities in the curricula of childcare centers in Daejeon area. 29 teachers from 21 day care centers were interviewed individually to obtain an understanding of teachers perceptions and conflicting views about play activity in the context of different day care curricula. The data were analysed qualitatively, using categorization and key word classification, frequency analyses and chi-squared tests. The results were as follows: (1) play-oriented curricula included sufficient play time and planned play activities in terms of the children`s development. Children`s freedom in play, optimum intercession by teachers, and interrelating activities between activity areas were included. In mixed curricula, children`s freedoms were limited in many cases, and interrelating play between activity areas was hardly found. Formal lesson-oriented curricula resulted unplanned play activities and teachers` passive intercession of play. (2) Most teachers believed that play activities were more important to a child`s development than formal lessons.

A survey of market orientation research (1995–2008)

Since 1990, market orientation (MO) has received considerable attention in the marketing literature. Little research so far provides preliminary and useful information to facilitate good understanding of MO. This study presents a comprehensive survey of extant marketing literature using keyword classification from 1995 to 2008. It serves as a tutorial and aims to help the beginner researcher or practitioner to have access to MO, including its definitions, applications and problem domains. Based on the scope of 514 MO articles, we find that the number of publications on MO has significantly increased since 2001 with steady growth in recent years. In addition, these selected articles are scattered across 153 journals and mostly published in 10 academic journals. The contribution of our paper is to provide a means (i.e., keyword index) to conceptualize and operationalize the coverage of MO. It provides a conceptual framework to organize this vast body of research. This study ends with some discussions and conclusions.

Using classification methods to label tasks in process mining

AbstractWe investigate a method designed to improve the accuracy of process mining in scenarios where the identification of task labels for log events is uncertain. Such situations are prevalent in business processes where events consist of communications between people, such as email messages. We examine how the accuracy of an independent task identifier, such as a classification or clustering engine, can be improved by examining the currently mined process model. First, a classification scheme based on identifying the keywords in each message is presented to provide an initial labeling. We then demonstrate how these labels can be refined by considering the likelihood that the event represents a particular task as obtained via an analysis of the current representation of the process model. This process is then repeated a number of times until the model is sufficiently refined. Results show that both keyword classification and the current process model analysis can be significantly effective on their own, and when combined have the potential to correct virtually all errors when noise is low (less than 20%), and can reduce the error rate by about 85% when noise is in the 30–40% range. Copyright © 2010 Crown in the right of Canada.

An intelligent agent-based system for multilingual financial news digest

Online financial news from different sources is widely available on the internet. There are systems available to help investors extract and analyse the financial news from these sources but many of these systems present news articles without categorisation and do not provide enough query options to accurately yet comprehensively search for news. In this paper, we extend our previous work to develop an intelligent agent-based system for multilingual news extraction. We adopt a document categorisation approach based on fuzzy keyword classification. The system applies fuzzy clustering to obtain a classification of keywords by concepts of the category. A category profile is developed and used as a search interface for document browsing. Experimental results show that the proposed categorise news agent is capable of categorising news documents with a reasonable rate of accuracy and the grouping news agent is able to assemble news groups of similar contents to facilitate information retrieval.

Artificial neural networks classification and clustering of methodologies and applications – literature analysis from 1995 to 2005

Abstract Based on a scope of 10,120 articles on ANNs, this paper uses data mining including association rules and cluster analysis, to survey these ANNs papers through keyword classification and clustering of articles from 1995 to 2005, exploring the ANNs methodologies and application developments during that period. The four decision variables of keywords, author’s nationality, research category, and year of publication, are implemented for data mining with total of 110,080 data items. The research findings show that some specific ANNs methodologies and applications pattern can be extracted from the mining results, and these describe the ANNs development over this period. In addition, using more data mining approaches for analysis could provide different explanations for ANNs development. Finally, discussion and brief conclusion are presented.

Automatic Document Classification

A computer cannot determine the subject content of a document the way the librarians do. Automatic classification is concerned with the procedures and systems that can make comparison between terms used. Using the computer to provide automatic indexes can only be considered for the automatic production of title indexes or the keyword enhanced index. Classificatory techniques, in fact, can be used for improving the efficiency of these title indexes. This paper attempts to trace solutions to the problems, such as truncation, stemming, etc. Automatic keyword classification is only an attempt. The use of automatically constructed classification is still a subject for investigation. http://dx.doi.org/10.14429/dbit.19.3.3486

Is the lack of keyword synergism inhibiting maturation in the MIS theory? An exploratory perspective

This article explores previous research on keyword search techniques and evaluates the comprehensiveness of a well-known keyword classification scheme. Comparison of literature reviews provide evidence that this topic is not well agreed upon and these contradictions have reduced reliance on keyword searches as a method of exploring prior research on a given topic. Presently, there does not exist a comprehensive management information systems (MIS) keyword list. Information regarding keyword usage in 14676 articles published in 27 MIS journals from 1985–1994 was collected electronically from an ABI Inform database. After several levels of refinements on 4386 keywords, a comprehensive list of 1056 keywords was obtained. Our MIS keyword list was compared with that of Barki and differences with Barki's list were identified, such as: (i) failing to include many keyword subjects that have frequently occurred during the past decade, and (ii) inclusion of keywords that have not had a single occurrence in MIS journals for the same period.

A comparative analysis of Reference Models for Visual and Computational Integrated Environments

In the framework of Visualization in Scientific Computing (ViSC) the definition of reference models appears to be an especially promising research area. In this paper we propose a unified approach to the analysis of reference models based on keyword classification. We analyze several models representing either visualization systems or complex computational and visual processes. We propose grouping the reference models in two sets called Reference Models in Scientific Visualization and Reference Models of Visual Computational Simulations respectively. For each of these two sets we present a general overview of reference models belonging to, we identify the main concepts and keywords characterizing the reference model definitions, and we present the result characterization organized in lists. We propose definitions of the structure of each of the reference model and representative keyword sets. To characterize specific models, and underline either their commonalties or distinctions, we label for instance two- or three-dimensional and one- or multilevel structures. We start to develop a comparative analysis of separate approaches and we identify elements significant for further advances in reference model design.

An Information Systems Keyword Classification Scheme

Information systems researchers, while helping others build better information systems, have done little in supporting their own information needs. The large amount of knowledge accumulated in IS today has created a situation where the researcher needs some sort of organizing mechanism to map the IS territory and relate its different components. A keyword classification scheme performs such a task through its content descriptors, systematically listed to show their relationships. This paper proposes a keyword classification scheme for IS. The scheme contains over 1,100 keywords and consists of nine top-level categories (reference disciplines, external environment, technological environment, organizational environment, IS management, IS development and operations, IS usage, information systems, and IS education and research), each of which is divided into subcategories.

What makes an automatic keyword classification effective?

AbstractThough the idea of constructing a keyword classification for retrieval purposes automatically is not a new one, comparatively few systematic experiments have been carried out in this area; and while many suggestions have been put forward, not enough is known about the behaviour of automatic keyword classifications, and hence about the properties such classifications should have and the ways they should be used. In previous experiments we showed that some forms of classification could give good results, and this paper describes a further series of tests designed to examine this sort of classification in more detail, with a view to establishing the optimum forms of classification and procedures for using them in different retrieval situations. These tests demonstrate that further improvements in performance over that for unclassified keywords can be obtained, and that definite conclusions can be drawn about the correct approach to classification for collections like the test one: the best results are given when grouping is confined to strongly connected, non‐frequent keywords, when the classification is used to provide additional rather than alternative indexing terms, particularly for requests, and when matching is controlled by keyword collection frequency.