Figure - uploaded by Geetha Tv
Content may be subject to copyright.
General purpose taxonomies (Wu et al., 2012)
Source publication
The prime textual sources used for ontology learning are a domain corpus and dynamic large text from web pages. The first source is limited and possibly outdated, while the second is uncertain. To overcome these shortcomings, a novel ontology learning methodology is proposed to utilize the different sources of text such as a corpus, web pages and t...
Context in source publication
Context 1
... overcome the aforesaid problems with information sources, we have used multi-sources of information such as a domain-specific corpus, web pages and a rich, universal and probabilistic taxonomy called Probase ( Wu et al., 2012). Although a handful of general-purpose taxonomies/ ontologies do exist (Table 1), they have limited concept space ( Wu et al., 2012) and hence Probase has been chosen. Probase consists of 2.7 million concepts (also multi-word terms) obtained from 1.8 billion web pages, and is constructed to handle the inconsistency, ambiguity and uncertainty of knowledge. ...Citations
... The origin of the knowledge utilized to create the ontology is referred to as the source of ontology. Due to the variety of knowledge sources, ontology is ideal for semantically representing knowledge by integrating and organizing it into a conceptual hierarchy [64]. ...
Knowledge management (KM) comprises several processes, and one of the most important is the knowledge sharing activities. The ability of an organization to manage its organizational knowledge, specifically in the context of knowledge sharing, may enhance the organization’s overall performance. Various approaches and technologies have been introduced to assist the process in achieving that target. Ontology as one of the knowledge representation methods has been becoming popular to assist knowledge sharing in the organization. Previous reviews have mainly focused on general KM issues, with little emphasis on the use of ontology in knowledge sharing. Thus, this article reviews several ontology-based KM tools that can support knowledge-sharing activities to provide some insight into future research in this area. Thirteen ontology-based KM tools were reviewed using ten elements’ comparison criteria: the motivation, domain, source of knowledge, type of knowledge, knowledge extraction, knowledge input process, knowledge retrieval process, knowledge sharing technology, source of ontology component, and ontology methodology. The review found that several elements can be further studied to improve KM implementation in the organization, especially on the knowledge sharing dimension. This includes simplifying the knowledge extraction and retrieval process to explore various knowledge domains from implicit knowledge sources. The review’s outcome also includes proposed components and functions of an ideal ontology-based KM tool.
... This kind of methods avoid the identification and modeling of diversified local and global characteristics. Therefore, compared with named entity recognition methods based on supervised learning, especially the methods based on deep learning, the target entity sample set and corresponding corpus set required by this kind of methods are smaller, which has been verified in a large number of previous research cases [15,34]. For examples, Roberto Navigl [28] et al. proposed the domain relevance and domain consensus (DR-DC) method, which calculated the domain relevance of candidate entities based on the word frequency and document frequency in target domain corpora and contrasting corpora. ...
Web farming can advance computational social science into a never-end learning process, in which social phenomena are dynamically and scientifically understood based on continuously produced, updated and expired data in the connected hyper world. Named entity recognition is a basic and core task of Web farming. However, the existing named entity recognition methods mainly depend on the complete, high-quality and well-labelled data sets and cannot meet the requirements of real-world applications. This paper proposes a continuous learning method for recognizing named entity by introducing the Web farming mode of Web Intelligence into the recognizing process. During the on-line stage, the domain contextual relevance of candidate entities is calculated by using the domain discrimination degree and the domain dependence function for recognizing the target entities. During the off-line stage, an active learning approach is designed to continuously improve the target corpus set by binding density-based clustering with semantic distance measurement. Experimental results show that the proposed method can effectively improve the accuracy of entity recognition and is more suitable for real-world applications.
... The case of definition ref.210 is the most striking since it comes as a direct quote. ...
Traditionally connected to Philosophy, the term ontology is increasingly related to information systems areas. Some researchers consider the approaches of the two disciplinary contexts to be completely different. Others consider that, although different, they should talk to each other, as both seek to answer similar questions. With the extensive literature on this topic, we intend to contribute to the understanding of the use of the term ontology in current research and which references support this use. An exploratory study was developed with a mixed methodology and a sample collected from the Web of Science of articles published in 2018. The results show the current prevalence of Computer Science in studies related to ontology and also of Gruber's view suggesting ontology as kind of conceptualization, a dominant view in that field. Some researchers, particularly in the field of Biomedicine, do not adhere to this dominant view, but to another one that seems closer to ontological study in the philosophical context. The term ontology, in the context of information systems, appears to be consolidating with a different meaning from the original presenting traces of the process of metaphorization in the transfer of the term between the two fields of study.
... The manual construction of ontologies for specific domains is a time-consuming and tedious task [6]. In contrast to manual ontology development, ontology learning (OL) aims to create ontologies automatically from given sources, such as textual and HTML documents or relational database (RDB) schema [7]. Thus, an OL approach helps reduce the time and effort consumed in ontology development. ...
Natural language processing discusses the applications of computational technique analysis and synthesis of natural languages. Semantic and morphological analysis are the two basic percepts in the natural language processing domain. Semantic analysis is the process of analyzing the lexical, grammatical, and syntactical parts of the words. The study of words known as morphology focuses on the meaning and structure of words. In this chapter, the authors focus on various morphological analyzers developed for Tamil language. Developing a highly accurate and adaptable morphological analyser is a challenging task. Morphological analyser basically identifies the morphemes and parts of speech for tagging. The atomic version of a word that retains the original meaning is called a morpheme. Morphological analyzer type includes phrase level and word level analyzers. Universal networking language (UNL) is a declarative kind used to express the natural language text using a semantic network. The major applications of UNL are information retrieval system, machine translation system, and UNL-based search engine.
This paper presents the learning strategy and the environment for Ontology Learning (OL) relations discovery task for the scientific publications domain by adjusting Deep Belief Network (DBN). The adjusted DBN is called Relu Dropout DBN (Re-DDBN). This paper elaborates on the adjusted Re-DDBN configuration, its structure, hyper-parameters, and functions. In addition, the adjusted Re-DDBN was compared with traditional DBN and other comparative models (e.g., Support vector machine (SVM) and Naïve Bayes (NB)) for ontology semantic relations discovery. The outcomes revealed that the adjusted Re-DDBN displayed the best performance when compared to the other models. The SemEval-2018 task 7 dataset was applied in this study.KeywordsActivation FunctionDeep belief NetworkDeep LearningDropout StrategyOntology LearningSemantic Relation
The banking industry performs credit score analysis as an efficient credit risk assessment method to determine a customer’s creditworthiness. In the banking industry, machine learning could be used for a variety of uses involving data analysis. A method of data analysis that is capable of self-regulation has been made possible by the development of modern techniques, such as classification approaches. The classification method is a form of supervised learning in which the computer acquires knowledge from the provided input data and then utilizes it to classify the dataset, which is used for training purposes. This study presents a comparative analysis of the various machine learning algorithms that are utilized to evaluate credit risk. The methods are used by utilizing the German Credit dataset that was collected from Kaggle, which consists of 1,000 instances and 11 attributes, all of which are used to determine if transactions are good or bad. The findings of data analysis using Logistic Regression, Linear Discriminant Analysis, Gaussian Naive Bayes, K-Nearest Neighbors Classifier, Decision Tree Classifier, Support Vector Machines, and Random Forest are compared and contrasted in this study. The findings demonstrated that the Random Forest algorithm forecasted credit risk effectively.KeywordsCredit RiskBankingMachine LearningPredictionFeatures
This paper presents the learning strategy and the environment for Ontology Learning (OL) relations discovery task for the scientific publications domain by adjusting Deep Belief Network (DBN). The adjusted DBN is called Relu Dropout DBN (Re-DDBN). This paper elaborates on the adjusted Re-DDBN configuration, its structure, hyper-parameters, and functions. In addition, the adjusted Re-DDBN was compared with traditional DBN and other comparative models (e.g., Support vector machine (SVM) and Naïve Bayes (NB)) for ontology semantic relations discovery. The outcomes revealed that the adjusted Re-DDBN displayed the best performance when compared to the other models. The SemEval-2018 task 7 dataset was applied in this study.
Different institutions have shown interest in standardizing the learning result. It may be used in the same way to assess students’ learning status. The teacher must quantify the learning outcomes for evaluation purposes. It often requires a great deal of time and effort to do paper tasks. Additionally, this activity prevents instructors from concentrating on the learning process. Teachers are continuously burdened with administrative responsibilities that should be alleviated using technology that adheres to the current framework. The Bloom Taxonomy, a widely used framework for defining learning outcomes, allows for the assessment of learning outcomes at several levels. The purpose of this research is to provide a framework that will assist the instructor in completing the evaluation more quickly and accurately. This study provided an algorithm for adapting ontology and text classification technologies to detect correlations between words and keywords to aid in evaluation. It is anticipated that the categorization findings will assist in shortening the time required to complete the evaluation.
Chapter 8
Augmented Reality based 3D Human Hands Tracking from Monocular True Images Using Convolutional Neural Network
A F M Saifuddin Saif, Zainal Rasyid Mahayuddin
Precise modeling of hand tracking from monocular camera calibration parameters using semantic cues is an active area of research for the researchers due to lack of accuracy and computational overheads. In this context, deep learning based framework, i.e. convolutional neural network based human hands tracking in the current camera frame become active research problem. In addition, tracking based on monocular camera needs to be addressed due to updated technology such as Unity3D engine and other related augmented reality plugins. This research aims to track human hands in continuous frame by using the tracked points to draw 3D model of the hands as an overlay. In the proposed methodology, Unity3D environment was used for localizing hand object in augmented reality (AR). Later, convolutional neural network was used to detect hand palm and hand keypoints based on cropped region of interest (ROI). Proposed method achieved accuracy rate of 99.2% where single monocular true images were used for tracking. Experimental validation shows the efficiency of the proposed methodology.
