Semantic Analysis of Text Data with Automated System

The reliability of quality management in industrial processes mainly depends on information about the traceability, precision and accuracy of a measurement system as well as on its systematic bias. The progressive development of networking and sensing in industrial machinery facilitates a quality-related process monitoring regarding information of measurement systems and singular sensor nodes. Hereby, integration on the information level is mandatory. Furthermore, information from the shop floor and from enterprise applications is needed to provide a consistent and integrated quality analysis. Thereby, these systems use different standards and technologies for exportation and propagation of data. Besides, integrative quality management and data analysis require enriched data that does not only comprise, for example, the measured value and its standard-dependent unit on the sensor level; rather, additional information is needed (e.g. the production process or the time and place of measurement). In this paper, a framework is presented that facilitates the semantic integration and analysis of measurement and enterprise data according to real-time requirements. Semantic technologies are used to encode the meaning of the data from the application code. Herewith, the data is automatically annotated using terms and concepts taken from the application domain. Furthermore, a semantic integration and transformation process is facilitated. Thus, subsequent integration and, most importantly, analysis processes can take advantage of these terms and concepts using specialized analysis algorithms. Besides, the conceptual application of the presented framework and processes in a high-pressure-die-casting scenario is presented.

ABSTRACTThe challenge of managing and extracting useful knowledge from social media data sources has attracted much attention from academics and industry. To address this challenge, semantic analysis of textual data is focused on in this paper. We propose an ontology-based approach to extract semantics of textual data and define the domain of data. In other words, we semantically analyze the social data at two levels: the entity level and the domain level. We have chosen Twitter as a social channel for the purpose of concept proof. Ontologies are used to capture domain knowledge and to enrich the semantics of tweets, by providing specific conceptual representation of entities that appear in the tweets. Case studies are used to demonstrate this approach. We experiment and evaluate our proposed approach with a public dataset collected from Twitter and from the politics domain. The ontology-based approach leverages entity extraction and concept mappings in terms of quantity and accuracy of concept identification.

This publication presents cognitive information systems dedicated to analysing and interpreting economic data. The systems for semantic data analysis presented here are exemplify UBMSS systems (Understanding-Based Management Support Systems). UBMSS systems are presented here using the example of two applications for assessing the liquidity and analysing the cash flow of a company.

This publication presents cognitive systems designed for analysing economic data. Such systems have been created as the next step in the development of classical DSS systems (Decision Support Systems), which are currently the most widespread tools providing computer support for economic decision-making. The increasing complexity of decision-making processes in business combined with increasing demands that managers put on IT tools supporting management cause DSS systems to evolve into intelligent information systems. This publication defines a new category of systems - UBMSS (Understanding Based Management Support Systems) which conduct in-depth analyses of data using on an apparatus for linguistic and meaning-based interpretation and reasoning. This type of interpretation and reasoning is inherent in the human way of perceiving the world. This is why the authors of this publication have striven to perfect the scope and depth of computer interpretation of economic information based on human processes of cognitive data analysis. As a result, they have created UBMSS systems for the automatic analysis and interpretation of economic data. The essence of the proposed approach to the cognitive analysis of economic data is the use of the apparatus for the linguistic description of data and for semantic analysis. This type of analysis is based on expectations generated automatically by a system which collects resources of expert knowledge, taking into account the information which can significantly characterise the analysed data. In this publication, the processes of classical data description and analysis are extended to include cognitive processes as well as reasoning and forecasting mechanisms. As a result of the analyses shown, we will present a new class of UBMSS cognitive economic information systems which automatically perform a semantic analysis of business data.

This paper presents a method for semantic data analysis by complex-valued matrix decomposition. The method is based on the quantum model of contextual decision-making, according to which observable probabilities are generated by the qubit states, representing the subjective meaning of the contexts relative to the basis decision. In the simplest three-context case, one of these qubits is decomposed into a superposition of the remaining two, mathematically encoding semantic relations between the three contexts. For use in data analysis, this model is translated to the matrix form, in which the rows and columns correspond to the contexts and samples of the experiment. The observable real-valued data then emerge from a complex-valued amplitude matrix, decomposed to a product of a real basis matrix and complex-valued matrix of superposition coefficients. This decomposition reveals stable process-semantic relations between the considered contexts, not captured by other methods of analysis. As a result, the data are approximated with higher precision and fewer parameters than singular and nonnegative matrix decompositions, truncated to the same dimension. The model is experimentally tested in the descriptive and prognostic regimes. The result opens prospects for development of nature-like computational architectures on novel logical grounds.

We propose an approach to the automatic categorization of text documents based on the joint application of the method of latent semantic analysis (LSA) and fuzzy inference Mamdani algorithm. Method LSA is used for the semantic analysis of information in electronic document management systems by identifying semantic relationships between terms of documents and receipt of the compliance rate of the compared vectors. The rule base is proposed for fuzzy inference algorithm of Mamdani implementing the automatic rubrication of documents for a variety of given topics enabling automated monitoring of the distribution of documents not relevant to the specified topics, or having similarities in several thematic categories on the basis of the results of latent semantic analysis.

An integrated approach for intrinsic plagiarism detection

In recent years, text mining has become a branch of Natural Language Processing. This chapter focuses on the analysis of textual data, illustrated by examples treated with the software R, which will be used to manipulate texts and strings, to transform them by word embedding methods and to analyse them by supervised and unsupervised statistical and machine learning methods. It discusses methods of topic modeling and sentiment analysis. There are two types of methods in textual data analysis: descriptive methods and predictive methods. The chapter describes the various lexical analyses that are often performed to categorize documents or to detect the topics contained in a set of documents. The analysis of textual data is most often done on vector representations of the data, which are stored in matrices containing the documents in rows and the terms in columns. A recent application of natural language processing is sentiment analysis.

Although massive open online courses (MOOCs) increase the number of choices in higher education and enhance learning, their low completion rate remains problematic. Previous studies have shown that learning engagement is a crucial factor influencing learning success and learner retention. However, few literature reviews on learning engagement in MOOCs have been conducted, and specific data analysis methods are lacking. Moreover, the internal and external factors that affect learning engagement have not been fully elucidated. Therefore, this systematic literature review summarized articles pertaining to learning engagement in MOOCs published from 2015 to 2022. Thirty articles met the inclusion and quality assurance criteria. We found that (1) learning engagement can be measured through analysis of log, text, image, interview, and survey data; (2) measures that have been used to analyze learning engagement include self-report (e.g., the Online Learning Engagement Scale, Online Student Engagement Questionnaire, and MOOC Engagement Scale) and automatic analysis methods [e.g., convolutional neural network (CNN), bidirectional encoder representations from transformers-CNN, K-means clustering, and semantic network analysis]; and (3) factors affecting learning engagement can be classified as internal (learning satisfaction, etc.) or external (curriculum design, etc.). Future research should obtain more diverse, multimodal data pertaining to social engagement. Second, researchers should employ automatic analysis methods to improve measurement accuracy. Finally, course instructors should provide technical support (“scaffolding”) for self-regulated learning to enhance student engagement with MOOCs.

Methods for analysis of principal components in discrete data have existed for some time under various names such as grade of membership modelling, probabilistic latent semantic analysis, and genotype inference with admixture. In this paper we explore a number of extensions to the common theory, and present some application of these methods to some common statistical tasks. We show that these methods can be interpreted as a discrete version of ICA. We develop a hierarchical version yielding components at different levels of detail, and additional techniques for Gibbs sampling. We compare the algorithms on a text prediction task using support vector machines, and to information retrieval.

Condition monitoring of complex systems is usually based on various sensors to gain information on the system status and identify potential problems at an early stage. However, evaluation and interpretation of multiple sensor data is often a major problem due to complex interdependencies between measured sensor data and actual system condition. We have addressed this problem for the example of a fluid condition monitoring system in wind turbines, i.e. an oil filter and cooling subsystem equipped with various sensors to monitor relevant temperatures, differential and absolute pressure drop across the filter as well as specific oil conditions such as dielectric constant and viscosity to monitor chemical oil degradation. Data were recorded from two wind turbines in a field test over more than two years. Data evaluation was performed with both statistical as well as semantic analysis methods with the goal of predicting the remaining filter lifetime and identifying unusual data patterns indicating potential problems for the overall system. Results indicate a large potential for the hybrid combination of statistical and semantic sensor data analysis for condition monitoring and have especially indicated the potential for learning data patterns with one specific system and transferring the results to other systems. This would allow using operational experience in larger networks such as offshore wind parks to improve the performance and reliability of intelligent condition monitoring systems.

A software architecture for Twitter collection, search and geolocation services

PurposeThe purpose of this paper is to apply the wavelet thresholding technique in order to analyze economic socio-political situations in Tunisia using textual data sets. This technique is used to remove noise from contingency table. A comparative study is done on correspondence analysis and classification results (using k-means algorithm) before and after denoising.Design/methodology/approachTextual data set is collected from an electronic newspaper that offers actual economic news about Tunisia. Both the hard and the soft-thresholding techniques are applied based on various Daubechies wavelets with different vanishing moments.FindingsThe results obtained have proved the effectiveness of wavelet denoising method in textual data analysis. On one hand, this technique allowed reducing the loss of information generated by correspondence analysis, ensured a better quality of representation of the factorial plan, neglected the interest of lemmatization in textual analysis and improved the results of classification by k-means algorithm. On the other hand, the proximities provided by the factorial visualization validate the economic situation of Tunisia during the studied period showing mainly a stable situation before the revolution and a deteriorated one after the revolution.Originality/valueThe results are the first to analyze economic socio-political relations using textual data. The originality of this paper comes also from the joint use of correspondence analysis and wavelet thresholding in textual data analysis.

In recent years, latent semantic analysis (LSA) has reached a level of maturity at which its presence is ubiquitous in technology as well as in simulation of cognitive processes. In spite of this, in recent years there has been a trend of subjecting LSA to some criticisms, usually because it is compared to other models in very specific tasks and conditions and sometimes without having good knowledge of what the semantic representation of LSA means, and without exploiting all the possibilities of which LSA is capable other than the cosine. This paper provides a critical review to clarify some of the misunderstandings regarding LSA and other space models. The historical stability of the predecessors of LSA, the representational structure of word meaning and the multiple topologies that could arise from a semantic space, the computation of similarity, the myth that LSA dimensions have no meaning, the computational and algorithm plausibility to account for meaning acquisition in LSA (in contrast to others models based on online mechanisms), the possibilities of spatial models to substantiate recent proposals, and, in general, the characteristics of classic vector models and their ease and flexibility to simulate some cognitive phenomena will be reviewed. The review highlights the similarity between LSA and other techniques and proposes using long LSA experiences in other models, especially in predicting models such as word2vec. In sum, it emphasizes the lessons that can be learned from comparing LSA-based models to other models, rather than making statements about "the best."

Within the context of investigating belief systems, this work exemplifies a mixed-method approach. Two approaches are utilized to explore the philosophical, ontological, and epistemological assumptions that university students foster regarding the relationships between knowledge and reality. In the first step, written materials that elaborated on the matter at hand were subjected to content analysis with the assistance of Leximancer, a software that recognizes themes and concepts and turns textual data into concept maps that express networks of meaning. The second step involved conducting a cluster analysis on the data obtained from the questionnaire to identify distinct groups of participants who shared consistent epistemological viewpoints. The results obtained from the two approaches are in agreement and shed light on the prevalent epistemic inclination that favors a constructivist viewpoint. Discussion on the ramifications of the findings, as well as the methodological issues that are pertinent to the present illustration, is provided.