Semantic Tags Research Articles

Automatic tagging by exploring tag information capability and correlation

Automatic tagging can automatically label images and videos with semantic tags to significantly facilitate multimedia search and organization. However, most of existing tagging algorithms often don't differentiate between tags used to describe visual content, and neglect the semantic correlation of the assigned tag set. In this paper, we propose a novel automatic tagging algorithm which tags a test image or video with an Informative and Correlative Tag (ICTag) set. The assigned ICTag set can provide a more precise description of the multimedia object by exploring both the information capability of individual tags and the tag-to-set correlation. Measures to effectively estimate the information capability of individual tags and the correlation between a tag and the candidate tag set are designed. To reduce the computational complexity, we also introduce a heuristic method to achieve efficient automatic tagging. We conduct extensive experiments on the NUS-WIDE web image dataset downloaded from Flickr and the MCG-WEBV web video dataset downloaded from YouTube. The results confirm the efficiency and effectiveness of our proposed algorithm.

Perspective Giving — Perspective Taking: Evidence-Based Learning in Organisations

Evidence for learning in organisations is given, once alternative or novel ways of task accomplishment are discovered by their stakeholders or management. Besides triggering a learning step, this knowledge should propagate to all concerned members of an organisation or a network of organisations, namely via context-sensitive exploration and re-construction of work processes. We introduce a learning framework based on existing organisational theories and findings. Its implementations cover generating evidence and interactive processing along individual and collective learning processes. Knowledge is supposed to be elicited in an intuitive way, either using a semantically open table top, or applying subject-oriented business-process modeling techniques. Keeping all provided inputs in an organisational memory allows stakeholders and management timely involvement in learning processes. The aligned Knowledge Management System features tracing of learning steps, sharing views, complementing expertise, and exploring content in a collaborative way. Implementation-wise evidence-based learning support does not only require semantic tagging and networking information, but also the exchangeability of perspectives on inputs or results of collaborative exploration sessions. Interactive enablers to that respect are annotations kept in views. Besides content enrichments, they contain links from the fine-grain content elements to social interactions, such as forum entries, as occurring in learning processes.

서지레코드와의 연계를 통한 폭소노미 태그 프레임워크 구축

현재의 정보환경에서 정보자원을 표현하는데 있어 여러 가지 방법들이 많이 적용되고 있다. 이 가운데 최근에는 태그를 이용한 폭소노미가 정보의 표현과 조직에 있어 새로운 방식으로 급격하게 대두되고 있다. 이는 기존의 방법들이 지닌 문제를 해결할 수 있는 방법으로서의 가능성을 지니고 있기는 하지만, 태그 자체가 지닌 모호성, 비일관성, 의미적 복합성 등으로 인해 효과적인 정보조직에 어려움을 가져오기도 한다. 이에 본 연구에서는 서지레코드와의 연계를 통해 태그의 의미적 제어를 위한 개념적 프레임워크를 구축하여 태그가 지닌 효율성을 극대화하고 태그의 고유한 문제를 해결하여 보다 효과적인 정보의 표현과 검색이 가능하도록 할 수 있는 방안을 제안하고 있다.

The Impact of Data Quality Tags on Decision-Making Outcomes and Process

It has been proposed that metadata describing data quality (DQ), termed DQ tags, be made available in situations where decision makers are unfamiliar with the data context, for example, in data warehouses. However, there have been conflicting reports as to the impact of such DQ tags on decision-making outcomes. Early studies did not explicitly consider the usability and semantics of the DQ tag designs used experimentally or the impact of such tags on decision process, except in suggestions for future research. This study addresses these issues, focusing on the design of usable DQ tags whose semantics are explicitly specified and exploring the impact of such DQ tags on decision outcomes and process. We use the information quality framework InfoQual, the interaction design technique of contextual inquiry, and cognitive process tracing to address DQ tag semantics, usability, and impact on decision process, respectively. In distinct contrast to earlier laboratory experiments, there was no evidence that the preferred decision choice changed with DQ tags, but decision time was significantly increased and there were indications of reduced consensus. These results can be explained by understanding the impact of DQ tags on decision process using concurrent protocol analysis, which involves participants verbalizing thoughts while making a decision. The protocol analysis study shows that DQ tags are associated with increased cognitive processing in the earlier phases of decision making, which delays generation of decision alternatives.

An intelligent content strategy for the enterprise

An intelligent content strategy for the enterprise

Semantic Tagging of Medical Narratives with Top Level Concepts from SNOMED CT Healthcare Data Standard

Medical narratives written by clinicians constitute critical information in healthcare domain and are required to be correct with respect to contextual meaning. SNOMED CT (Systematized Nomenclature of Medicine -- Clinical Terms) is a standardized reference terminology that consists of 390023 SNOMED CT concepts with SNOMED CT codes. This paper describes the extraction of SNOMED CT concepts from free text discharge summary reports. For the evaluation of the medical concepts, we used 300 discharge summaries corpus provided by University of Pittsburgh Medical Centre, and compared it with the SNOMED CT concept file which is a preprocessed and cleaned file listing SNOMED CT concepts. In this paper we present the ongoing research on SNOMED CT concept extraction from discharge summaries using natural language processing and introducing SNOMED CT core concepts as a single gazetteer list for concept extraction. Out of 390023 concepts, 21563 concepts were found in the test set of discharge summaries from the SNOMED CT core concepts gazetteer list. A modified approach extracted the 23 top level concept tags which will be useful for linguist analysis research.

The Semantic Tagging Model of Chinese Question Chunk Based on Description Logic

To solve the problem of Chinese question cannot provide deep semantic information for syntactic analysis, the paper proposes a semantic tagging model of Chinese question chunk based on description logics. With the knowledge of HNC’s concept symbols to get semantic information about the words, and by starting with the connotation of the concept categories, it preliminarily analyzes the logic structure of specific question sentences. Finally, with description logics reasoning mechanism, we get question semantic view and practical verify the validity of it.

Survey on social tagging techniques

Social tagging on online portals has become a trend now. It has emerged as one of the best ways of associating metadata with web objects. With the increase in the kinds of web objects becoming available, collaborative tagging of such objects is also developing along new dimensions. This popularity has led to a vast literature on social tagging. In this survey paper, we would like to summarize different techniques employed to study various aspects of tagging. Broadly, we would discuss about properties of tag streams, tagging models, tag semantics, generating recommendations using tags, visualizations of tags, applications of tags and problems associated with tagging usage. We would discuss topics like why people tag, what influences the choice of tags, how to model the tagging process, kinds of tags, different power laws observed in tagging domain, how tags are created, how to choose the right tags for recommendation, etc. We conclude with thoughts on future work in the area.

Toward an integrative software infrastructure for water management in the Smarter Planet

Building a Smarter Planet™ requires creating an intelligent infrastructure that integrates technology with business, government, and the everyday life of the citizens of earth, to maximize the use of scarce resources, balance human use and ecosystem preservation, reduce costs, and improve quality of life. One of the keystones of this intelligent infrastructure is an integrative modeling framework (IMF), which is a platform for enabling the integration by nonexpert users of diverse sensor-based data, related business data, and complex cross-disciplinary mathematical modeling, in support of planning, monitoring, management, reporting, and decision support applications. We describe a research prototype that applies the Mashup Automation with Runtime Invocation and Orchestration (MARIO) technology from IBM Research to this problem in a specific application area in water management: simulating stream discharges using compositions of hydrologic process submodels derived from monolithic stream discharge simulators. We show how MARIO's semantic tagging and model composition engine enable us to meet three critical challenges of an IMF: 1) generating valid chains or compositions of model components, given a definition of starting and ending states; 2) allowing all scientifically valid compositions of components; and 3) disallowing compositions that are scientifically invalid, i.e., that combine model components whose basic assumptions about quantities such as soil architectures or evaporation schemes conflict. While we focus here on water management, the technology that we describe can readily be generalized to other intelligent infrastructure applications, e.g., cities, transportation, and utilities.

Using automated semantic tagging in Critical Discourse Analysis: A case study on Scottish independence from a Scottish nationalist perspective

To date, studies of social attitudes towards Scottish independence tend to have been of the structured survey or interview variety. This study seeks to support and build upon the findings of recent social attitude surveys on Scottish independence using what is, as far as the author is aware, a novel methodology. This involves combining the corpus linguistic technique of automated semantic tagging with a discourse-historical Critical Discourse Analysis (CDA) framework.Applying this to a three-million-word corpus built from a pro-independence internet discussion forum, the analysis shows, firstly, a view that independence will strengthen, consolidate or transform Scottish identity in a positive way and, secondly, a distinct lack of strategies that seek to dismantle British identity or refer to historical disputes. Thus, an evaluation of this methodology suggests that it successfully manages to produce findings that support previous research, challenge existing stereotypes, and allow new insight into Scottish nationalist ideology.

Semantic tagging of and semantic enhancements to systematics papers: ZooKeys working examples

The concept of semantic tagging and its potential for semantic enhancements to taxonomic papers is outlined and illustrated by four exemplar papers published in the present issue of ZooKeys. The four papers were created in different ways: (i) written in Microsoft Word and submitted as non-tagged manuscript (doi: 10.3897/zookeys.50.504); (ii) generated from Scratchpads and submitted as XML-tagged manuscripts (doi: 10.3897/zookeys.50.505 and doi: 10.3897/zookeys.50.506); (iii) generated from an author’s database (doi: 10.3897/zookeys.50.485) and submitted as XML-tagged manuscript. XML tagging and semantic enhancements were implemented during the editorial process of ZooKeys using the Pensoft Mark Up Tool (PMT), specially designed for this purpose. The XML schema used was TaxPub, an extension to the Document Type Definitions (DTD) of the US National Library of Medicine Journal Archiving and Interchange Tag Suite (NLM). The following innovative methods of tagging, layout, publishing and disseminating the content were tested and implemented within the ZooKeys editorial workflow: (1) highly automated, fine-grained XML tagging based on TaxPub; (2) final XML output of the paper validated against the NLM DTD for archiving in PubMedCentral; (3) bibliographic metadata embedded in the PDF through XMP (Extensible Metadata Platform); (4) PDF uploaded after publication to the Biodiversity Heritage Library (BHL); (5) taxon treatments supplied through XML to Plazi; (6) semantically enhanced HTML version of the paper encompassing numerous internal and external links and linkouts, such as: (i) vizualisation of main tag elements within the text (e.g., taxon names, taxon treatments, localities, etc.); (ii) internal cross-linking between paper sections, citations, references, tables, and figures; (iii) mapping of localities listed in the whole paper or within separate taxon treatments; (v) taxon names autotagged, dynamically mapped and linked through the Pensoft Taxon Profile (PTP) to large international database services and indexers such as Global Biodiversity Information Facility (GBIF), National Center for Biotechnology Information (NCBI), Barcode of Life (BOLD), Encyclopedia of Life (EOL), ZooBank, Wikipedia, Wikispecies, Wikimedia, and others; (vi) GenBank accession numbers autotagged and linked to NCBI; (vii) external links of taxon names to references in PubMed, Google Scholar, Biodiversity Heritage Library and other sources. With the launching of the working example, ZooKeys becomes the first taxonomic journal to provide a complete XML-based editorial, publication and dissemination workflow implemented as a routine and cost-efficient practice. It is anticipated that XML-based workflow will also soon be implemented in botany through PhytoKeys, a forthcoming partner journal of ZooKeys. The semantic markup and enhancements are expected to greatly extend and accelerate the way taxonomic information is published, disseminated and used.

Taxonomy shifts up a gear: New publishing tools to accelerate biodiversity research

Biodiversity is under severe threat. It is estimated that there are untold millions of species on the planet and we have, in the past 250 years and recently with a decreasing number of specialists, described only about 2 million of them.Th ough this happens with an agonizing low description rate of an estimated number of 20,000 new species per year (Polaszek et al. 2005), digital and genomic resources now are allowing discovery of new and identifi cation of existing species at a never before seen pace. Th e principles of Open Access greatly facilitate dissemination of information through the Web where it is freely accessed, shared and updated in a form that is accessible to indexing and data mining engines using Web 2.0 technologies. Web 2.0 turns the taxonomic information into a global resource well beyond the taxonomic community. A signifi cant bottleneck in naming species is the requirement by the current Codes of biological nomenclature ruling that new names and their associated descriptions must be published on paper, which can be slow, costly and render the new information diffi cult to fi nd. In order to make progress in documenting the diversity of life, we must remove the publishing impediment in order to move taxonomy “from a cottage industry into a production line” (Lane et al. 2008), and to make best use of new technologies warranting the fastest and widest distribution of these new results. In this special edition of ZooKeys we present a practical demonstration of such a process. Th e issue opens with a forum paper from Penev et al. (doi: 10.3897/zookeys.50.538) that presents the landscape of semantic tagging and text enhancements in taxonomy. It describes how the content of the manuscript is enriched by semantic tagging and marking up of four exemplar papers submitted to the publisher in three diff erent ways: (i) written in Microsoft Word and submitted as non-tagged manuscript (Stoev et al., doi: 10.3897/zookeys.50.504); (ii) generated from Scratchpads (Blagoderov et al., doi: 10.3897/zookeys.50.506 and Brake and Tschirnhaus, doi: 10.3897/ zookeys.50.505); (iii) generated from an author’s database (Taekul et al., doi: 10.3897/ zookeys.50.485). Th e latter two were submitted as XML-tagged manuscript. Th ese examples demonstrate the suitability of the workfl ow to a range of possibilities that should encompass most current taxonomic eff orts. To implement the aforementioned routes for XML mark up in prospective taxonomic publishing, a special software tool ZooKeys 50: i-iv (2010)

Social media filtering based on collaborative tagging in semantic space

We propose a semantic collaborative filtering method to enhance recommendation quality derived from user-generated tags. Social tagging is employed as an approach in order to grasp and filter users' preferences for items. In addition, we explore several advantages of semantic tagging for ambiguity, synonymy, and semantic interoperability, which are notable challenges in information filtering. The proposed approach first determines semantically similar users using social tagging and subsequently discovers semantically relevant items for each user. Experimental results show that our method offers significant advantages both in terms of improving the recommendation quality and in dealing with ambiguity, synonymy, and interoperability issues.

What Do I Want from the Publisher of the Future?

When I took on the role of Editor-in-Chief of this open-access journal, I began,for the first time, to think about scholarlycommunication beyond submitting my pa-pers and getting them published. Thisthinking led to previous Perspectives [1–3],all of which shared an underlying theme—there are many opportunities to achievebetter dissemination and comprehension ofour science, and as producers of that outputI believe authors have a responsibility to seeit used in the best possible way.No need to take my word regarding theopportunities that exist to improve schol-arly communication and comprehension. Irecommend reading ‘‘Part 4: ScholarlyCommunication’’ from the free onlinebook the Fourth Paradigm: Data Intensive Scien-tific Discovery [4] (http://research.microsoft.com/en-us/collaboration/fourthparadigm/),which is a tribute to the late Turing Awardwinner Jim Gray. Jim, and many of theauthors who pay homage to his vision, havethought deeply about the subject of scholarlycommunication. They conclude that dataand knowledge-driven computation is indeeda fourth wave, as computation has impactedscience to the point where every aspect of itis touched by computation (hence the nameeScience), including dissemination and com-prehension. These visionaries recognize thatwe are at a tipping point at which scholarlycommunication will change from a tradi-tional print-oriented medium (albeit an on-line version of the print journal) to somethingelse. That something else begins to transformtoday’s research article as we realize thepower of the medium, establish new forms ofknowledge discovery, and measure theimpact of scholarly contributions in newways. For all that vision, these luminaries donot address the question that I have beenpondering, and which I would like to raisehere. Assuming all this innovation takesplace, what will the publisher of the futurelook like, and as a contributor and consumerof a publisher’s services in this new era, whatdo I want from the publisher of the future?Recently, at gatherings of publisherswhere I have been invited to speak, I havebeen trying to pose and then answer thisquestion. Unfortunately, I fear that what Ipropose appears so radical as to be greetedwitheitherblankstaresorlooksof getreal.LetmetryheretodoabetterjobatstatingwhatI want from my publisher in the future.Many of you are undoubtedly thinkingthat just accepting your papers will beenough, but bear with me. Presumably,publishing will continue in the life sciences(unless we go over completely to anArXiv.org or similar model where articlesare simply deposited without peer reviewand impact measured by how much theyare accessed), and if so, will continue to beoverseen by the publishers we, as scientists,work with today. A few new and innova-tive publishers like the Public Library ofScience (PLoS) will continue to emerge asbusiness models and practices change, butexisting publishers will probably adapt inthis new era. I anticipate similarities toearlier phases of the Internet revolution.Amazon.com emerged as a new and majoronline-only shopping entity, but Sears,Wal-Mart, Harrods, etc., while beingslower in adopting the new medium, dideventually successfully support onlineshopping and a range of new services. Bycomparison, a few innovators have hadsome impact on scholarly communication,but traditional science, technology, andmedical (STM) publishers will continue todominate the conservative and relativelyslow-moving market. These pioneeringpublishers are now experimenting withinteractive PDFs, ‘‘articles of the future,’’semantic tagging, data integration withresearch articles, incorporating rich media(video and podcasts), and so on. Mostlikely, at some point these innovations willbecome mainstream through increasedintroduction by traditional publishers, butthen what? Stated another way, if wefinally move away from the traditionalPDF to something more dynamic thatintegrates data, rich media, and includesinteractive access, what do I as a scientistwant from publishers at that point?To answer this question, let us start withwhere we are today. As authors, we put anenormous amount of effort into producingapublishablemanuscript.Atsomepointwepass it over to the publisher without asecondthought.Subsequently,wewillputalarge amount of effort into a revision orrebuttal letter, but again, there is nothought on what will happen to our workafterithasbeenacceptedbeyondthedateitwill be published and appear in PubMed.There is an enormous amount of trust inour publisher that our creations will behandled in the best possible way and, whenpublished, that they will be disseminated toallwhowanttoreadourwork.Openaccessintroduced a hairline fracture in this trustwith some scientists realizing that perhapstheir work was notbeingaswidely accessedaspossible.Nevertheless,mostscientistsstilldo not think seriously about limited accessand signing away the copyright. After alloureffortsatproducingapaper,veryfewofus have asked the question, is journal xpresenting my work in a way that maxi-mizes the understanding of what has beendone, providing the means to ensuremaximum reproducibility of what has beendone, and maximizing the outreach of mywork? I would suggest that now is the timenottojusttossthepaperoverahighbarrierto the journal and forget about it, but tobreak down the barrier and have a newform of interaction and dialog with apublisher who is prepared to embrace achanging publishing model and cananswerthe question in a satisfactory manner. Inother words, we have an interaction withthe publisher that does not begin when thescientific process ends, but begins at thebeginning of the scientific process itself.Perhaps you are beginning to see why Iget so many blank stares when I raise thisissue with scientists (producers and consum-

Why do Users Tag? Detecting Users’ Motivation for Tagging in Social Tagging Systems

While recent progress has been achieved in understanding the structure and dynamics of social tagging systems, we know little about the underlying user motivations for tagging, and how they influence resulting folksonomies and tags. This paper addresses three issues related to this question: 1.) What motivates users to tag resources, and in what ways is user motivation amenable to quantitative analysis? 2.) Does users' motivation for tagging vary within and across social tagging systems, and if so how? and 3.) How does variability in user motivation influence resulting tags and folksonomies? In this paper, we present measures to detect whether a tagger is primarily motivated by categorizing or describing resources, and apply the measures to datasets from 8 different tagging systems. Our results show that a) users' motivation for tagging varies not only across, but also within tagging systems, and that b) tag agreement among users who are motivated by categorizing resources is significantly lower than among users who are motivated by describing resources. Our findings are relevant for (i) the development of tag recommenders, (ii) the analysis of tag semantics and (iii) the design of search algorithms for social tagging systems.

Supporting Ontology-Based Semantic Annotation of Business Processes with Automated Suggestions

Annotation of Business Processes with semantic tags taken from a domain ontology is beneficial to several activities conducted on Business Processes, such as comprehension, documentation, analysis and evolution. On the other hand, the task of semantically annotating Business Processes is time-consuming and far from trivial. The authors support Business Process designers in the annotation of process elements by automatically suggesting candidate concepts. The annotation suggestions are computed on the basis of a similarity measure between the text information associated with process element labels and the ontology concepts. In turn, this requires support for the disambiguation of terms appearing in ontology concepts, which admit multiple linguistic senses, and for ontology extension, when the available concepts are insufficient.

Web Information Extraction by Semantic Tagging

An important aspect of research for Web information extraction relates to the inference of complex reasoning and correlation based on distributed information available in many different Web data sources. By defining the semantics of information and services available on the Web, the World Wide Web becomes a vast store of information that can be easily processed by computer applications.

Will Widgets and Semantic Tagging Change Computational Biology?

We argue here, through the use of several examples from our work in support of structural biology, that the answer to the question posed by the title of this Perspective is a resounding yes. The discussion that follows is aimed primarily at those of the journal's readers who are biological resource developers and Web page developers interested in developing the richest possible Web pages. However, those of you who simply use biological resources might find this a helpful discussion in understanding what is on the horizon. Whatever your interest, please let us hear your opinion on the question posed by this Perspective through the associated comment feature.

Semantic APIs: Scaling up towards the Semantic Web

Web information retrieval and knowledge discovery are undergoing changes. The size of the Web and the heterogeneity of web pages generate new challenges in meeting user needs. This paper investigates the different methods deployed that add semantics to web content: semantic tagging and semantic APIs. The research carried out investigates existing systems in each category, outlining their primary features and functionality. It then proposes a framework for the evaluation of semantic tagging based on the main requirements for information discovery and recommends a number of comparative assessments, ranging from basic product information and requirements’ analysis to the evaluation of the APIs information modelling functionality.

An Ontological Framework for Retrieving Environmental Sounds Using Semantics and Acoustic Content

Organizing a database of user-contributed environmental sound recordings allows sound files to be linked not only by the semantic tags and labels applied to them, but also to other sounds with similar acoustic characteristics. Of paramount importance in navigating these databases are the problems of retrieving similar sounds using text- or sound-based queries, and automatically annotating unlabeled sounds. We propose an integrated system, which can be used for text-based retrieval of unlabeled audio, content-based query-by-example, and automatic annotation of unlabeled sound files. To this end, we introduce an ontological framework where sounds are connected to each other based on the similarity between acoustic features specifically adapted to environmental sounds, while semantic tags and sounds are connected through link weights that are optimized based on user-provided tags. Furthermore, tags are linked to each other through a measure of semantic similarity, which allows for efficient incorporation of out-of-vocabulary tags, that is, tags that do not yet exist in the database. Results on two freely available databases of environmental sounds contributed and labeled by nonexpert users demonstrate effective recall, precision, and average precision scores for both the text-based retrieval and annotation tasks.