Information

Research

15 pages, 342 KiB

Open AccessArticle

Comparing Supervised Machine Learning Strategies and Linguistic Features to Search for Very Negative Opinions

by Sattam Almatarneh and Pablo Gamallo

Information 2019, 10(1), 16; https://doi.org/10.3390/info10010016 - 04 Jan 2019

Cited by 7 | Viewed by 4055

In this paper, we examine the performance of several classifiers in the process of searching for very negative opinions. More precisely, we do an empirical study that analyzes the influence of three types of linguistic features (n-grams, word embeddings, and polarity lexicons) and [...] Read more.

In this paper, we examine the performance of several classifiers in the process of searching for very negative opinions. More precisely, we do an empirical study that analyzes the influence of three types of linguistic features (n-grams, word embeddings, and polarity lexicons) and their combinations when they are used to feed different supervised machine learning classifiers: Naive Bayes (NB), Decision Tree (DT), and Support Vector Machine (SVM). The experiments we have carried out show that SVM clearly outperforms NB and DT in all datasets by taking into account all features individually as well as their combinations. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

15 pages, 996 KiB

Open AccessArticle

Semantic Distance Spreading Across Entities in Linked Open Data

by Sultan Alfarhood, Susan Gauch and Kevin Labille

Information 2019, 10(1), 15; https://doi.org/10.3390/info10010015 - 02 Jan 2019

Cited by 4 | Viewed by 2877

Abstract

Recommender systems can utilize Linked Open Data (LOD) to overcome some challenges, such as the item cold start problem, as well as the problem of explaining the recommendation. There are several techniques in exploiting LOD in recommender systems; one approach, called Linked Data [...] Read more.

Recommender systems can utilize Linked Open Data (LOD) to overcome some challenges, such as the item cold start problem, as well as the problem of explaining the recommendation. There are several techniques in exploiting LOD in recommender systems; one approach, called Linked Data Semantic Distance (LDSD), considers nearby resources to be recommended by calculating a semantic distance between resources. The LDSD approach, however, has some drawbacks such as its inability to measure the semantic distance resources that are not directly linked to each other. In this paper, we first propose another variation of the LDSD approach, called wtLDSD, by extending indirect distance calculations to include the effect of multiple links of differing properties within LOD, while prioritizing link properties. Next, we introduce an approach that broadens the coverage of LDSD-based approaches beyond resources that are more than two links apart. Our experimental results show that approaches we propose improve the accuracy of the LOD-based recommendations over our baselines. Furthermore, the results show that the propagation of semantic distance calculation to reflect resources further away in the LOD graph extends the coverage of LOD-based recommender systems. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

23 pages, 719 KiB

Open AccessArticle

A Comparison of Word Embeddings and N-gram Models for DBpedia Type and Invalid Entity Detection

by Hanqing Zhou, Amal Zouaq and Diana Inkpen

Information 2019, 10(1), 6; https://doi.org/10.3390/info10010006 - 25 Dec 2018

Cited by 3 | Viewed by 3739

Abstract

This article presents and evaluates a method for the detection of DBpedia types and entities that can be used for knowledge base completion and maintenance. This method compares entity embeddings with traditional N-gram models coupled with clustering and classification. We tackle two challenges: [...] Read more.

This article presents and evaluates a method for the detection of DBpedia types and entities that can be used for knowledge base completion and maintenance. This method compares entity embeddings with traditional N-gram models coupled with clustering and classification. We tackle two challenges: (a) the detection of entity types, which can be used to detect invalid DBpedia types and assign DBpedia types for type-less entities; and (b) the detection of invalid entities in the resource description of a DBpedia entity. Our results show that entity embeddings outperform n-gram models for type and entity detection and can contribute to the improvement of DBpedia’s quality, maintenance, and evolution. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

24 pages, 908 KiB

Open AccessArticle

Improving the Accuracy in Sentiment Classification in the Light of Modelling the Latent Semantic Relations

by Nina Rizun, Yurii Taranenko and Wojciech Waloszek

Information 2018, 9(12), 307; https://doi.org/10.3390/info9120307 - 04 Dec 2018

Cited by 9 | Viewed by 3546

Abstract

The research presents the methodology of improving the accuracy in sentiment classification in the light of modelling the latent semantic relations (LSR). The objective of this methodology is to find ways of eliminating the limitations of the discriminant and probabilistic methods for LSR [...] Read more.

The research presents the methodology of improving the accuracy in sentiment classification in the light of modelling the latent semantic relations (LSR). The objective of this methodology is to find ways of eliminating the limitations of the discriminant and probabilistic methods for LSR revealing and customizing the sentiment classification process (SCP) to the more accurate recognition of text tonality. This objective was achieved by providing the possibility of the joint usage of the following methods: (1) retrieval and recognition of the hierarchical semantic structure of the text and (2) development of the hierarchical contextually-oriented sentiment dictionary in order to perform the context-sensitive SCP. The main scientific contribution of this research is the set of the following approaches: at the phase of LSR revealing (1) combination of the discriminant and probabilistic models while applying the rules of adjustments to obtain the final joint result; at all SCP phases (2) considering document as a complex structure of topically completed textual components (paragraphs) and (3) taking into account the features of persuasive documents’ type. The experimental results have demonstrated the enhancement of the SCP accuracy, namely significant increase of average values of recall and precision indicators and guarantee of sufficient accuracy level. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

12 pages, 690 KiB

Open AccessArticle

Ontology-Based Representation for Accessible OpenCourseWare Systems

by Mirette Elias, Steffen Lohmann and Sören Auer

Information 2018, 9(12), 302; https://doi.org/10.3390/info9120302 - 29 Nov 2018

Cited by 9 | Viewed by 3589

Abstract

OpenCourseWare (OCW) systems have been established to provide open educational resources that are accessible by anyone, including learners with special accessibility needs and preferences. We need to find a formal and interoperable way to describe these preferences in order to use them in [...] Read more.

OpenCourseWare (OCW) systems have been established to provide open educational resources that are accessible by anyone, including learners with special accessibility needs and preferences. We need to find a formal and interoperable way to describe these preferences in order to use them in OCW systems and retrieve relevant educational resources. This formal representation should use standard accessibility definitions of OCW that can be reused by other OCW systems to represent accessibility concepts. In this article, we present an ontology to represent the accessibility needs of learners with respect to the IMS AfA specifications. The ontology definitions together with rule-based queries are used to retrieve relevant educational resources. Related to this, we developed a user interface component that enables users to create accessibility profiles representing their individual needs and preferences based on our ontology. We evaluated the approach with five examples profiles. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

20 pages, 1022 KiB

Open AccessArticle

Alignment: A Hybrid, Interactive and Collaborative Ontology and Entity Matching Service

by Sotirios Karampatakis, Charalampos Bratsas, Ondřej Zamazal, Panagiotis Marios Filippidis and Ioannis Antoniou

Information 2018, 9(11), 281; https://doi.org/10.3390/info9110281 - 15 Nov 2018

Cited by 6 | Viewed by 3819

Abstract

Ontology matching is an essential problem in the world of Semantic Web and other distributed, open world applications. Heterogeneity occurs as a result of diversity in tools, knowledge, habits, language, interests and usually the level of detail. Automated applications have been developed, implementing [...] Read more.

Ontology matching is an essential problem in the world of Semantic Web and other distributed, open world applications. Heterogeneity occurs as a result of diversity in tools, knowledge, habits, language, interests and usually the level of detail. Automated applications have been developed, implementing diverse aligning techniques and similarity measures, with outstanding performance. However, there are use cases where automated linking fails and there must be involvement of the human factor in order to create, or not create, a link. In this paper we present Alignment, a collaborative, system aided, interactive ontology matching platform. Alignment offers a user-friendly environment for matching two ontologies with the aid of configurable similarity algorithms. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

28 pages, 2490 KiB

Open AccessArticle

Smart Process Optimization and Adaptive Execution with Semantic Services in Cloud Manufacturing ^†

by Luca Mazzola, Philipp Waibel, Patrick Kaphanke and Matthias Klusch

Information 2018, 9(11), 279; https://doi.org/10.3390/info9110279 - 13 Nov 2018

Cited by 15 | Viewed by 5070

Abstract

A new requirement for the manufacturing companies in Industry 4.0 is to be flexible with respect to changes in demands, requiring them to react rapidly and efficiently on the production capacities. Together with the trend to use Service-Oriented Architectures (SOA), this requirement induces [...] Read more.

A new requirement for the manufacturing companies in Industry 4.0 is to be flexible with respect to changes in demands, requiring them to react rapidly and efficiently on the production capacities. Together with the trend to use Service-Oriented Architectures (SOA), this requirement induces a need for agile collaboration among supply chain partners, but also between different divisions or branches of the same company. In order to address this collaboration challenge, we propose a novel pragmatic approach for the process analysis, implementation and execution. This is achieved through sets of semantic annotations of business process models encoded into BPMN 2.0 extensions. Building blocks for such manufacturing processes are the individual available services, which are also semantically annotated according to the Everything-as-a-Service (XaaS) principles and stored into a common marketplace. The optimization of such manufacturing processes combines pattern-based semantic composition of services with their non-functional aspects. This is achieved by means of Quality-of-Service (QoS)-based Constraint Optimization Problem (COP) solving, resulting in an automatic implementation of service-based manufacturing processes. The produced solution is mapped back to the BPMN 2.0 standard formalism by means of the introduced extension elements, fully detailing the enactable optimal process service plan produced. This approach allows enacting a process instance, using just-in-time service leasing, allocation of resources and dynamic replanning in the case of failures. This proposition provides the best compromise between external visibility, control and flexibility. In this way, it provides an optimal approach for business process models’ implementation, with a full service-oriented taste, by implementing user-defined QoS metrics, just-in-time execution and basic dynamic repairing capabilities. This paper presents the described approach and the technical architecture and depicts one initial industrial application in the manufacturing domain of aluminum forging for bicycle hull body forming, where the advantages stemming from the main capabilities of this approach are sketched. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

12 pages, 375 KiB

Open AccessArticle

Reducing the Deterioration of Sentiment Analysis Results Due to the Time Impact

by Yuliya Rubtsova

Information 2018, 9(8), 184; https://doi.org/10.3390/info9080184 - 25 Jul 2018

Cited by 8 | Viewed by 3587

Abstract

The research identifies and substantiates the problem of quality deterioration in the sentiment classification of text collections identical in composition and characteristics, but staggered over time. It is shown that the quality of sentiment classification can drop up to 15% in terms of [...] Read more.

The research identifies and substantiates the problem of quality deterioration in the sentiment classification of text collections identical in composition and characteristics, but staggered over time. It is shown that the quality of sentiment classification can drop up to 15% in terms of the F-measure over a year and a half. This paper presents three different approaches to improving text classification by sentiment in continuously-updated text collections in Russian: using a weighing scheme with linear computational complexity, adding lexicons of emotional vocabulary to the feature space and distributed word representation. All methods are compared, and it is shown which method is most applicable in certain cases. Experiments comparing the methods on sufficiently representative text collections are described. It is shown that suggested approaches could reduce the deterioration of sentiment classification results for collections staggered over time. Full article

(This article belongs to the Special Issue Knowledge Engineering and Semantic Web)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Knowledge Engineering and Semantic Web

Share This Special Issue

Special Issue Editors

Special Issue Information

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI