Semantic-based Techniques Research Articles

A Novel Framework for the Generation of Multiple Choice Question Stems Using Semantic and Machine-Learning Techniques

Multiple Choice Questions (MCQs) are a popular assessment method because they enable automated evaluation, flexible administration and use with huge groups. Despite these benefits, the manual construction of MCQs is challenging, time-consuming and error-prone. This is because each MCQ is comprised of a question called the "stem", a correct option called the "key" along with alternative options called "distractors" whose construction demands expertise from the MCQ developers. In addition, there are different kinds of MCQs such as Wh-type, Fill-in-the-blank, Odd one out, and many more needed to assess understanding at different cognitive levels. Automatic Question Generation (AQG) for developing heterogeneous MCQ stems has generally followed two approaches: semantics-based and machine-learning-based. Questions generated via AQG techniques can be utilized only if they are grammatically correct. Semantics-based techniques have been able to generate a range of different types of grammatically correct MCQs but require the semantics to be specified. In contrast, most machine-learning approaches have been primarily able to generate only grammatically correct Fill-in-the-blank/Cloze by reusing the original text. This paper describes a technique for combining semantic-based and machine-learning-based techniques to generate grammatically correct MCQ stems of various types for a technical domain. Expert evaluation of the resultant MCQ stems demonstrated that they were promising in terms of their usefulness and grammatical correctness.

Verification of Distributed Systems via Sequential Emulation

Sequential emulation is a semantics-based technique to automatically reduce property checking of distributed systems to the analysis of sequential programs. An automated procedure takes as input a formal specification of a distributed system, a property of interest, and the structural operational semantics of the specification language and generates a sequential program whose execution traces emulate the possible evolutions of the considered system. The problem as to whether the property of interest holds for the system can then be expressed either as a reachability or as a termination query on the program. This allows to immediately adapt mature verification techniques developed for general-purpose languages to domain-specific languages, and to effortlessly integrate new techniques as soon as they become available. We test our approach on a selection of concurrent systems originated from different contexts from population protocols to models of flocking behaviour. By combining a comprehensive range of program verification techniques, from traditional symbolic execution to modern inductive-based methods such as property-directed reachability, we are able to draw consistent and correct verification verdicts for the considered systems.

A forensic-driven data model for automatic vehicles events analysis

Digital vision technologies emerged exponentially in all living areas to watch, play, control, or track events. Security checkpoints have benefited also from those technologies by integrating dedicated cameras in studied locations. The aim is to manage the vehicles accessing the inspection security point and fetching for any suspected ones. However, the gathered data volume continuously increases each day, making their analysis very hard and time-consuming. This paper uses semantic-based techniques to model the data flow between the cameras, checkpoints, and administrators. It uses ontologies to deal with the increased data size and its automatic analysis. It considers forensics requirements throughout the creation of the ontology modules to ensure the records’ admissibility for any possible investigation purposes. Ontology-based data modeling will help in the automatic events search and correlation to track suspicious vehicles efficiently.

Use of a domain-specific ontology to support automated document categorization at the concept level: Method development and evaluation

Voluminous, conveniently accessible textual documents, created and disseminated by modern information technology, makes automated document organization increasingly important for both individuals and organizations. Many existing techniques rely on document content analysis that classifies new, unlabeled documents by examining the similarity based on the overlap between their important features and the representative features of each document category. However, the performance of feature-based techniques can be significantly hindered by word mismatch and ambiguity problems. As a remedy, this study takes a concept-based approach and propose a text categorization method that incorporates a domain-specific ontology to support automated document categorization more effectively. The proposed method classifies documents according to their respective range of relevant concepts. We empirically evaluate our method versus several prevalent benchmarks that include feature-based k-nearest neighbors (kNN) and semantic-based techniques. The results show the proposed method more effective than the benchmark techniques; it achieves better performances when using a complete concept hierarchy without considering the hierarchical relationships among concepts. The proposed method illustrates how to incorporate a domain-specific ontology to improve document classification. Our method is computationally efficient because it produces a concept space of relatively few dimensionalities and does not require semantic space reconstruction as new documents arrive. Moreover, the relationships and patterns for classifying documents, generated by our method, are explicit and comprehensible.

Performance Exploration on Various Document Clustering Techniques with K-Means Family

Clustering performs a important position in numerous fields which include Computer science & packages, facts, pattern reputation, system studying technique and find out dating among the files. Clustering focuses on document clustering, and other related area. Increase within the extent of statistics saved in virtual form (text, photograph, audio) has improved the need for requirement of an automated tool, that allows people to find and manage the records in an efficient way. Usually clustering refer to document clustering technique investigates the documents and find its relation. This paper center of attention on the a range of clustering methods and evaluation its overall performance. This paper also categories the document clustering techniques as three major groups, namely Group K-means, Expectation Maximization and Semantic-based techniques (Hybrid method). Several experiments were conducted to analyse the performance accuracy and Speed.

The Dynamic Web Services Adaptation Framework in Context-Aware Mobile Cloud Learning Using Semantic-Based Method

Most of Service-Based Applications (SBAs) have to be changed after their first deployment not solely due to the changing system requirements as well as of continuous change of the environment itself. With the growth of web service paradigm, there is a need for an efficient mechanism in dynamic adaptation process to offer users a better service experience. In particular, context-aware can be adapted in dynamic adaptation by considering user's contexts and device's contexts. Context awareness will be a key area for Mobile Cloud Learning (MCL) environment as it helps in the reasoning process to provide correct education resources for learners' prospect. Contextual information are represented using semantic-based approach for high expressiveness and formal representation for reasoning technique. However, dynamic adaptations frameworks in previous research are still lacking in terms of contextual information of the learner and device, and quality of services (QoS) were not considered. Besides, there is limited support of semantic expressiveness in terms of contextual information and services by using solely semantic-based technique. This paper proposes Dynamic Adaptation in Context Aware Mobile Cloud Learning (DACAMoL) framework to support adaptation process in MCL using semantic-based approach. The framework is applied in MCL mobile application that offers basic learning language.

Analysis and shortcomings of e-recruitment systems: Towards a semantics-based approach addressing knowledge incompleteness and limited domain coverage

The rapid development of the Internet has led to introducing new methods for e-recruitment and human resources management. These methods aim to systematically address the limitations of conventional recruitment procedures through incorporating natural language processing tools and semantics-based methods. In this context, for a given job post, applicant resumes (usually uploaded as free-text unstructured documents in different formats such as .pdf, .doc or .rtf) are matched/screened out using the conventional keyword-based model enriched by additional resources such as occupational categories and semantics-based techniques. Employing these techniques has proved to be effective in reducing the cost, time, and efforts required in traditional recruitment and candidate selection methods. However, bridging the skill gap - that is, the propensity to precisely detect and extract relevant skills in applicant resumes and job posts - and highlighting the hidden semantic dimensions encoded in applicant resumes are still challenging issues in the process of devising effective e-recruitment systems. This is due to the fact that resources exploited by current e-recruitment systems are obtained from generic domain-independent sources, therefore resulting in knowledge incompleteness and the lack of domain coverage. In this article, we review state-of-the-art e-recruitment approaches and highlight recent advancements in this domain. An e-recruitment framework addressing current shortcomings through the use of multiple cooperative semantic resources, feature extraction techniques and skill relatedness measures is detailed. An instantiation of the proposed framework is proposed and an experimental validation using a real-world recruitment dataset from two employment portals demonstrates the effectiveness of the proposed approach.

Creative design of emergency management scenarios driven by semantics: An application to smart cities

We present a framework to support creative design of emergency management scenarios. By creative design of scenarios we mean the process of imagining situations and describing them through models and stories. The framework supports the tasks of gathering and organizing knowledge about emergency management situations by automatically generating conceptual models, related to fragments of emergency scenarios. It leverages semantics-based techniques to enable a computational creativity approach. A software application was defined to support the activities of modeling scenarios by permitting to generate, organize, and query sets of these conceptual models, which we name mini-stories, and that can be adopted to inspire the activity of creative design. Selected mini-stories are blueprints for more detailed user scenario descriptions and models that can be used, for instance, for analysis or simulation. As a case study, we consider emergency management in smart cities. This is a challenging domain because smart cities are characterized by interconnected physical and virtual services forming complex ecosystems, which provide sophisticated services to the population and to institutions, manage public resources in a optimal way, and involve citizens in decisional processes. As a consequence, smart city ecosystems can be threatened by several hazards spanning from natural disasters, as tsunami and earthquakes, to anthropic events, as terrorist attacks. Ability of service providers and institutional operators to face and manage emergency situations is therefore a relevant issue. Simulation and analysis of both crisis events and executions of management plans are a promising approach to deal with these articulated problems. However, manual definition of models to base the analysis is a demanding activity due to the huge number of different situations to consider. It requires knowledge related to the crisis and emergency domains, to the context (e.g., a specific city and its current regulations) and ability in modeling tasks. All these aspects demand for tools to support modeling activities, and our proposal aims at fulfilling this need. In particular, the discussed framework uses in a integrated way three types of knowledge: structural knowledge, to support the construction of models based on design patterns; domain knowledge, here related to smart cities and emergency management and represented by means of ontologies; and contextual knowledge, related to specific aspects (e.g., localization) of the considered scenario and represented as rules. We validated the presented approach by means of experiments performed by real city planners.

An assessment on domain ontology-based information extraction techniques

Domain ontology is used in information retrieval to retrieve more relevant information from a collection of unstructured information source. Information retrieval is becoming an important research area in the field of computer science. Information retrieval (IR) is generally concerned with the searching and retrieving of knowledge-based information from database. In this paper, we represent the various models and techniques for information retrieval. Domain ontology is a description of domain concepts with relation and properties to be used in knowledge engineering as a knowledge base. In this paper, various domain ontology-based information retrieval methods have been reviewed. A comparative analysis is made on all the available methods, which will allow the analyst to choose the suitable domain ontology-based information extraction method. There are various methods developed to make the information extraction more efficient. The methods have been classified as Boolean, vector space, semantic-based techniques and probabilistic. Semantic-based information retrieval can still be classified as semantic association, semantic similarity and semantic annotation. This assessment allows the developer to choose the best fit model for their requirement in an efficient way.

An assessment on domain ontology-based information extraction techniques

Domain ontology is used in information retrieval to retrieve more relevant information from a collection of unstructured information source. Information retrieval is becoming an important research area in the field of computer science. Information retrieval (IR) is generally concerned with the searching and retrieving of knowledge-based information from database. In this paper, we represent the various models and techniques for information retrieval. Domain ontology is a description of domain concepts with relation and properties to be used in knowledge engineering as a knowledge base. In this paper, various domain ontology-based information retrieval methods have been reviewed. A comparative analysis is made on all the available methods, which will allow the analyst to choose the suitable domain ontology-based information extraction method. There are various methods developed to make the information extraction more efficient. The methods have been classified as Boolean, vector space, semantic-based techniques and probabilistic. Semantic-based information retrieval can still be classified as semantic association, semantic similarity and semantic annotation. This assessment allows the developer to choose the best fit model for their requirement in an efficient way.

Formal Specification and Verification of Few Combined Fragments of UML Sequence Diagram

UML has become a de-facto standard for design and development of object-oriented systems. On the other hand, UML includes various diagrams and notations which are missing of formal semantics. There does not exist much work to transform completely UML models to executable codes by any of the computer-aided software engineering tools because of the hidden semantics under the UML diagrams. Moreover, UML models cannot be used for automated analysis for validation and verification of the systems. Therefore, an integration of UML and semantics-based techniques such as formal methods is required to overcome such type of issues. Further, proof of correctness of software models plays a key role to ensure the quality of the final system to be developed. This paper presents a systematic method for transformation and verification of UML sequence diagrams into Z specification. The method works by translating and analysing few important combined fragments of sequence diagrams using Z notation. At first, formal specification of primary constructs is provided by capturing the hidden semantics under the diagram. Formal analysis of interaction operators namely options, alternatives, loops and break of the fragments is provided. The combined fragments are important because it allows multiple scenarios to increase an expressive power of sequence diagram by various control flows. Finally, formal analysis is provided by Z/Eves tool to check the syntax, type and proof for consistency and correctness of the specification.

Ontology-Based Analysis of Microarray Data.

The importance of semantic-based methods and algorithms for the analysis and management of biological data is growing for two main reasons. From a biological side, knowledge contained in ontologies is more and more accurate and complete, from a computational side, recent algorithms are using in a valuable way such knowledge. Here we focus on semantic-based management and analysis of protein interaction networks referring to all the approaches of analysis of protein-protein interaction data that uses knowledge encoded into biological ontologies. Semantic approaches for studying high-throughput data have been largely used in the past to mine genomic and expression data. Recently, the emergence of network approaches for investigating molecular machineries has stimulated in a parallel way the introduction of semantic-based techniques for analysis and management of network data. The application of these computational approaches to the study of microarray data can broad the application scenario of them and simultaneously can help the understanding of disease development and progress.

Generalized semantic Web service composition

With the increasing popularity of Web Services and Service-Oriented Architecture, we need infrastructure to discover and compose Web services. In this paper, we present a generalized semantics-based technique for automatic service composition that combines the rigor of process-oriented composition with the descriptiveness of semantics. Our generalized approach presented in this paper introduces the use of a conditional directed acyclic graph where complex interactions, containing control flow, information flow, and pre-/post-conditions are effectively represented. Composition solution obtained is represented semantically as OWL-S documents. Web service composition will gain wider acceptance only when users know that the solutions obtained are comprised of trustworthy services. We present a framework that not only uses functional and non-functional attributes provided by the Web service description document but also filters and ranks solutions based on their trust rating that is computed using Centrality Measure of Social Networks. Our contributions are applied for automatic workflow generation in context of the currently important bioinformatics domain. We evaluate our engine for automatic workflow generation of a phylogenetic inference task. We also evaluate our engine for automated discovery and composition on repositories of different sizes and present the results.

The Aesop Approach for Semantic-Based End-User Service Optimization

The need to autonomically optimize end-user service experience in near real time has been identified in the literature in recent years. Management systems that monitor end-user service session context exist but approaches that estimate end-user service experience from session context do not analyze the compliance of that experience with user expectations. Approaches that optimize end-user service delivery are not applicable to arbitrary services; they either optimize specific service types or use general mechanisms that do not consider service experience. The lack of a holistic model for end-user service management is a barrier to autonomic end-user service optimization. This paper presents Aesop, an approach addressing autonomic optimization of end-user service delivery using semantic-based techniques. Its knowledge base uses the End-User Service Analysis and Optimization ontology, which models the end-user service management domain and partitions knowledge that varies over time for efficient access. The Aesop Engine executes an autonomic loop in near real time, which runs semantic algorithms to monitor sessions, analyze their compliance with expectations, and plan and execute optimizations on service delivery networks. The algorithms are efficient because they operate on small partitioned subsets of the Knowledge Base held as separate self-contained models at run time. An Aesop implementation was evaluated on a home area network test bed where compliance of service sessions with expectations when optimization was active was compared with compliance of an identical set of sessions when optimization was inactive. Significant improvements were observed on compliance levels of high priority sessions in all experimental scenarios, with compliance levels more than doubled in some cases.

A Flexible Architecture for Service Management in the Cloud

Cloud computing is a style of computing where different capabilities are provided as a service to customers using Internet technologies. The most common offered services are Infrastructure (IasS), Software (SaaS) and Platform (PaaS). This work integrates the service management into the cloud computing concept and shows how management can be provided as a service in the cloud. Nowadays, services need to adapt their functionalities across heterogeneous environments with different technological and administrative domains. The implied complexity of this situation can be simplified by a service management architecture in the cloud. This paper focuses on this architecture, taking into account specific service management functionalities, like incident management or KPI/SLA management, and provides a complete solution. The proposed architecture is based on a distributed set of agents, using semantic-based techniques: a Shared Knowledge Plane, instantiated in the cloud, has been introduced to ensure communication between agents.

Semantics Driven Approach for Knowledge Acquisition From EMRs

Semantic computing technologies have matured to be applicable to many critical domains such as national security, life sciences, and health care. However, the key to their success is the availability of a rich domain knowledge base. The creation and refinement of domain knowledge bases pose difficult challenges. The existing knowledge bases in the health care domain are rich in taxonomic relationships, but they lack nontaxonomic (domain) relationships. In this paper, we describe a semiautomatic technique for enriching existing domain knowledge bases with causal relationships gleaned from Electronic Medical Records (EMR) data. We determine missing causal relationships between domain concepts by validating domain knowledge against EMR data sources and leveraging semantic-based techniques to derive plausible relationships that can rectify knowledge gaps. Our evaluation demonstrates that semantic techniques can be employed to improve the efficiency of knowledge acquisition.

PREDOSE: A semantic web platform for drug abuse epidemiology using social media

ObjectivesThe role of social media in biomedical knowledge mining, including clinical, medical and healthcare informatics, prescription drug abuse epidemiology and drug pharmacology, has become increasingly significant in recent years. Social media offers opportunities for people to share opinions and experiences freely in online communities, which may contribute information beyond the knowledge of domain professionals. This paper describes the development of a novel semantic web platform called PREDOSE (PREscription Drug abuse Online Surveillance and Epidemiology), which is designed to facilitate the epidemiologic study of prescription (and related) drug abuse practices using social media. PREDOSE uses web forum posts and domain knowledge, modeled in a manually created Drug Abuse Ontology (DAO – pronounced dow), to facilitate the extraction of semantic information from User Generated Content (UGC), through combination of lexical, pattern-based and semantics-based techniques. In a previous study, PREDOSE was used to obtain the datasets from which new knowledge in drug abuse research was derived. Here, we report on various platform enhancements, including an updated DAO, new components for relationship and triple extraction, and tools for content analysis, trend detection and emerging patterns exploration, which enhance the capabilities of the PREDOSE platform. Given these enhancements, PREDOSE is now more equipped to impact drug abuse research by alleviating traditional labor-intensive content analysis tasks. MethodsUsing custom web crawlers that scrape UGC from publicly available web forums, PREDOSE first automates the collection of web-based social media content for subsequent semantic annotation. The annotation scheme is modeled in the DAO, and includes domain specific knowledge such as prescription (and related) drugs, methods of preparation, side effects, and routes of administration. The DAO is also used to help recognize three types of data, namely: (1) entities, (2) relationships and (3) triples. PREDOSE then uses a combination of lexical and semantic-based techniques to extract entities and relationships from the scraped content, and a top-down approach for triple extraction that uses patterns expressed in the DAO. In addition, PREDOSE uses publicly available lexicons to identify initial sentiment expressions in text, and then a probabilistic optimization algorithm (from related research) to extract the final sentiment expressions. Together, these techniques enable the capture of fine-grained semantic information, which facilitate search, trend analysis and overall content analysis using social media on prescription drug abuse. Moreover, extracted data are also made available to domain experts for the creation of training and test sets for use in evaluation and refinements in information extraction techniques. ResultsA recent evaluation of the information extraction techniques applied in the PREDOSE platform indicates 85% precision and 72% recall in entity identification, on a manually created gold standard dataset. In another study, PREDOSE achieved 36% precision in relationship identification and 33% precision in triple extraction, through manual evaluation by domain experts. Given the complexity of the relationship and triple extraction tasks and the abstruse nature of social media texts, we interpret these as favorable initial results. Extracted semantic information is currently in use in an online discovery support system, by prescription drug abuse researchers at the Center for Interventions, Treatment and Addictions Research (CITAR) at Wright State University. ConclusionA comprehensive platform for entity, relationship, triple and sentiment extraction from such abstruse texts has never been developed for drug abuse research. PREDOSE has already demonstrated the importance of mining social media by providing data from which new findings in drug abuse research were uncovered. Given the recent platform enhancements, including the refined DAO, components for relationship and triple extraction, and tools for content, trend and emerging pattern analysis, it is expected that PREDOSE will play a significant role in advancing drug abuse epidemiology in future.

Typestate-based semantic code search over partial programs

We present a novel code search approach for answering queries focused on API-usage with code showing how the API should be used. To construct a search index, we develop new techniques for statically mining and consolidating temporal API specifications from code snippets. In contrast to existing semantic-based techniques, our approach handles partial programs in the form of code snippets. Handling snippets allows us to consume code from various sources such as parts of open source projects, educational resources (e.g. tutorials), and expert code sites. To handle code snippets, our approach (i) extracts a possibly partial temporal specification from each snippet using a relatively precise static analysis tracking a generalized notion of typestate, and (ii) consolidates the partial temporal specifications, combining consistent partial information to yield consolidated temporal specifications, each of which captures a full(er) usage scenario. To answer a search query, we define a notion of relaxed inclusion matching a query against temporal specifications and their corresponding code snippets. We have implemented our approach in a tool called PRIME and applied it to search for API usage of several challenging APIs. PRIME was able to analyze and consolidate thousands of snippets per tested API, and our results indicate that the combination of a relatively precise analysis and consolidation allowed PRIME to answer challenging queries effectively.

A Comparative Study To Find A Suitable Method For Text Document Clustering

Text mining is used in various text related tasks such as information extraction, concept/entity extraction, document summarization, entity relation modeling (i.e., learning relations between named entities), categorization/classification and clustering. This paper focuses on document clustering, a field of text mining, which groups a set of documents into a list of meaningful categories. The main focus of this paper is to present a performance analysis of various techniques available for document clustering. The results of this comparative study can be used to improve existing text data mining frameworks and improve the way of knowledge discovery. This paper considers six clustering techniques for document clustering. The techniques are grouped into three groups namely Group 1 - K-means and its variants (traditional K-means and K* Means algorithms), Group 2 - Expectation Maximization and its variants (traditional EM, Spherical Gaussian EM algorithm and Linear Partitioning and Reallocation clustering (LPR) using EM algorithms), Group 3 - Semantic-based techniques (Hybrid method and Feature-based algorithms). A total of seven algorithms are considered and were selected based on their popularity in the text mining field. Several experiments were conducted to analyze the performance of the algorithm and to select the winner in terms of cluster purity, clustering accuracy and speed of clustering.

Going beyond completeness in information retrieval

The importance of information (or content) is becoming higher as computers penetrate everyday life. This has speeded up information management researches. We focus on information retrieval in this paper and present a classification of semantic-based techniques which rely on semantic aspects of people and surroundings as well as data objects themselves. We also present the notion of cognition-based retrieval which counts our cognitive nature in retrieval processing as a new promising approach. What this means is that, as similar to the theory of relativity introduced by Einstein, information spaces should form a continuum which bends, folds or warps from the observer's point of view.