• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.2
• /
• pp.471-492
• /
• 2017

The demand and interest in big data analytics are increasing rapidly. The concepts around big data include not only existing structured data, but also various kinds of unstructured data such as text, images, videos, and logs. Among the various types of unstructured data, text data have gained particular attention because it is the most representative method to describe and deliver information. Text analysis is generally performed in the following order: document collection, parsing and filtering, structuring, frequency analysis, and similarity analysis. The results of the analysis can be displayed through word cloud, word network, topic modeling, document classification, and semantic analysis. Notably, there is an increasing demand to identify trending topics from the rapidly increasing text data generated through various social media. Thus, research on and applications of topic modeling have been actively carried out in various fields since topic modeling is able to extract the core topics from a huge amount of unstructured text documents and provide the document groups for each different topic. In this paper, we review the major techniques and research trends of text analysis. Further, we also introduce some cases of applications that solve the problems in various fields by using topic modeling.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.455-460
• /
• 2023

Text augmentation is a methodology that creates new augmented texts by transforming or generating original texts for the purpose of improving the performance of NLP models. However existing text augmentation techniques have limitations such as lack of expressive diversity semantic distortion and limited number of augmented texts. Recently text augmentation using large language models and few-shot learning can overcome these limitations but there is also a risk of noise generation due to incorrect generation. In this paper, we propose a text augmentation method called TAGS that generates multiple candidate texts and selects the appropriate text as the augmented text. TAGS generates various expressions using few-shot learning while effectively selecting suitable data even with a small amount of original text by using contrastive learning and similarity comparison. We applied this method to task-oriented chatbot data and achieved more than sixty times quantitative improvement. We also analyzed the generated texts to confirm that they produced semantically and expressively diverse texts compared to the original texts. Moreover, we trained and evaluated a classification model using the augmented texts and showed that it improved the performance by more than 0.1915, confirming that it helps to improve the actual model performance.

• Asia pacific journal of information systems
• /
• v.18 no.1
• /
• pp.79-96
• /
• 2008

One of the roles of the Semantic Web services is to execute dynamic intra-organizational services including the integration and interoperation of business processes. Since different organizations design their processes differently, the retrieval of similar semantic business processes is necessary in order to support inter-organizational collaborations. Most approaches for finding services that have certain features and support certain business processes have relied on some type of logical reasoning and exact matching. This paper presents our approach of using imprecise matching for expanding results from an exact matching engine to query the OWL(Web Ontology Language) MIT Process Handbook. MIT Process Handbook is an electronic repository of best-practice business processes. The Handbook is intended to help people: (1) redesigning organizational processes, (2) inventing new processes, and (3) sharing ideas about organizational practices. In order to use the MIT Process Handbook for process retrieval experiments, we had to export it into an OWL-based format. We model the Process Handbook meta-model in OWL and export the processes in the Handbook as instances of the meta-model. Next, we need to find a sizable number of queries and their corresponding correct answers in the Process Handbook. Many previous studies devised artificial dataset composed of randomly generated numbers without real meaning and used subjective ratings for correct answers and similarity values between processes. To generate a semantic-preserving test data set, we create 20 variants for each target process that are syntactically different but semantically equivalent using mutation operators. These variants represent the correct answers of the target process. We devise diverse similarity algorithms based on values of process attributes and structures of business processes. We use simple similarity algorithms for text retrieval such as TF-IDF and Levenshtein edit distance to devise our approaches, and utilize tree edit distance measure because semantic processes are appeared to have a graph structure. Also, we design similarity algorithms considering similarity of process structure such as part process, goal, and exception. Since we can identify relationships between semantic process and its subcomponents, this information can be utilized for calculating similarities between processes. Dice's coefficient and Jaccard similarity measures are utilized to calculate portion of overlaps between processes in diverse ways. We perform retrieval experiments to compare the performance of the devised similarity algorithms. We measure the retrieval performance in terms of precision, recall and F measure? the harmonic mean of precision and recall. The tree edit distance shows the poorest performance in terms of all measures. TF-IDF and the method incorporating TF-IDF measure and Levenshtein edit distance show better performances than other devised methods. These two measures are focused on similarity between name and descriptions of process. In addition, we calculate rank correlation coefficient, Kendall's tau b, between the number of process mutations and ranking of similarity values among the mutation sets. In this experiment, similarity measures based on process structure, such as Dice's, Jaccard, and derivatives of these measures, show greater coefficient than measures based on values of process attributes. However, the Lev-TFIDF-JaccardAll measure considering process structure and attributes' values together shows reasonably better performances in these two experiments. For retrieving semantic process, we can think that it's better to consider diverse aspects of process similarity such as process structure and values of process attributes. We generate semantic process data and its dataset for retrieval experiment from MIT Process Handbook repository. We suggest imprecise query algorithms that expand retrieval results from exact matching engine such as SPARQL, and compare the retrieval performances of the similarity algorithms. For the limitations and future work, we need to perform experiments with other dataset from other domain. And, since there are many similarity values from diverse measures, we may find better ways to identify relevant processes by applying these values simultaneously.

• Journal of the Korean Society for information Management
• /
• v.22 no.3 s.57
• /
• pp.261-287
• /
• 2005

The purpose of this paper is to explore the ways to improve the performance of SVM (Support Vector Machines) text classifier using inter-document similarities. SVMs are powerful machine learning systems, which are considered as the state-of-the-art technique for automatic document classification. In this paper text categorization via SVMs approach based on feature representation with document vectors is suggested. In this approach, document vectors instead of index terms are used as features, and vector similarities instead of term weights are used as feature values. Experiments show that SVM classifier with document vector features can improve the document classification performance. For the sake of run-time efficiency, two methods are developed: One is to select document vector features, and the other is to use category centroid vector features instead. Experiments on these two methods show that we can get improved performance with small vector feature set than the performance of conventional methods with index term features.

• Journal of Korean Elementary Science Education
• /
• v.33 no.4
• /
• pp.674-682
• /
• 2014

This study aims to analyze STEAM-related articles and to look into the trend of research to present implications for research directions in the future. To achieve the research purpose, the researcher searched by key words, 'STEAM' and 'Convergence Education' through the RISS. Subjects of analysis were titles of 181 articles in journal articles and conference papers published from 2011 through 2013. Through an analysis of the frequency of the texts that appeared in the titles of the papers, key words were selected, the co-occurrence matrix of the key words was established, and using network maps, degree centrality and betweenness centrality, and structural equivalence, a network text analysis was carried out. For the analysis, KrKwic, KrTitle, UCINET and NetMiner Program were used, and the results were as follows: in the result of the text frequency analysis, the key words appeared in order of 'program', 'development', 'base' and 'application'. Through the network among the texts, a network built up with core hubs such as 'program', 'development', 'elementary' and 'application' was found, and in the degree centrality analysis, 'program', 'elementary', 'development' and 'science' comprised key issues at a relatively high value, which constituted the pivot of the network. As a result of the structural equivalence analysis, regarding the types of their respective relations, it was analyzed that there was a similarity in four clusters such as the development of a program (1), analysis of effects (2) and the establishment of a theoretical base (1).

• Journal of English Language & Literature
• /
• v.58 no.5
• /
• pp.859-880
• /
• 2012

This paper aims at analyzing Robert Southey's Thalaba the Destroyer in relation to cultural colonialism of the British Romantic period and investigating the ways in which this text portrays the Other through its literary representation of the East. Especially, this paper attempts to show that the Oriental world constructed in Southey's text reveals the imperial subject's self-conscious awareness of its unstable relation with the unknown Other. For this purpose, this paper attends to the formal aspects of Thalaba the Destroyer, examining the process by which the reader's generic expectations about the "epic" undergo complex revisions and frustrations through reading this text. The epic elements contained in Thalaba the Detroyer include the battle between good and evil and the hero's moral epiphany arising from his struggle against malicious enemies. Yet, Thalaba the Destroyer constantly destabilizes the distinction between self and other by leading the reader to recognize the uncomfortable similarity between the poem's tyrannical figures and imperialistic monarchs in the Western civilization. Thus, when the hero enacts a revolution against despotism, the resistant power points not only to the imagined false kingdom within the text, but to the core of the real Empire that seeks to construct its own "garden" in the global scene. In addition, Southey's "panoramic" description of Oriental objects and stories in his footnotes lacks a framing perspective, erasing and de-stabilizing subject/object distinctions. In these footnotes, he exposes his profound attraction to the culture of "Other" and also conveys his aspiration to transforming Eastern myths and stories into profitable literary texts. Southey's attitude to the East in the footnotes appears to be partially grounded upon the interest of mercantile capitalists of the West, who need to discover potential commodities. Yet, simultaneously, he reveals a sense of moral hesitation about his own desire for the materiality of the East, along with deep anxiety arising from the fear of punishment.

• Journal of Korea Entertainment Industry Association
• /
• v.5 no.1
• /
• pp.127-133
• /
• 2011

Characters in natural image are an important information with various context. Previous work of character region detection algorithms is not detect of character region in case of image complexity and the surrounding lighting, similar background to character, so this paper propose an method of character region detection in natural image using edge and connected component by morphological reconstructions. Firstly, we detect edge using Canny-edge detector and connected component with local min/max value by morphological reconstructed-operation in gray-scale image, and labeling each of detected connected component elements. lastly, detected candidate of text regions was merged for generation for one candidate text region, Final text region detected by checking the similarity and adjacency of neighbor of text candidate individual character. As the results of experiments, proposed algorithm improved the correctness of character regions detection using edge and connected components.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.728-743
• /
• 2014

Finger printing and sequence alignment are well-known approaches for document similarity comparison. A fingerprinting method is simple and fast, but it can not find particular similar regions. A string alignment method is used for identifying regions of similarity by arranging the sequences of a string. It has an advantage of finding particular similar regions, but it also has a disadvantage of taking more computing time. The Multi-Level Alignment (MLA) is a new method designed for taking the advantages of both methods. The MLA divides input documents into uniform length blocks, and then extracts fingerprints from each block and calculates similarity of block pairs by comparing the fingerprints. A similarity table is created in this process. Finally, sequence alignment is used for specifying longest similar regions in the similarity table. The MLA allows users to change block's size to control proportion of the fingerprint algorithm and the sequence alignment. As a document is divided into several blocks, similar regions are also fragmented into two or more blocks. To solve this fragmentation problem, we proposed a united block method. Experimentally, we show that computing document's similarity with the united block is more accurate than the original MLA method, with minor time loss.

• Journal of Digital Convergence
• /
• v.10 no.5
• /
• pp.267-273
• /
• 2012

The semantic relationship structures hierarchically the huge amount of document objects which is usually not formatted. However, it is very difficult to structure relevant data from various distributed application domains. This paper proposed a new object management method to service the distributed domain objects by using semantic reference relationship. The proposed mechanism utilized the profile structure in order to extract the semantic similarity from application domain objects and utilized the joint matrix to decide the semantic relationship of the extracted objects. This paper performed the simulation to show the performance of the proposed method, and simulation results show that the proposed method has better retrieval performance than the existing text mining method and information extraction method.

• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.2
• /
• pp.150-161
• /
• 2015

This paper proposes a new approach to technology valuation, the market-replacement cost approach which integrates the cost-based approach and market-based approach. The proposed approach estimates the market-replacement cost of a target technology using R&D costs of similar R&D projects previously conducted. Similar R&D projects are extracted from project database based on document similarity between project proposals and technology description of the target technology. R&D costs of similar R&D projects are adjusted by mirroring the rate of technological obsolescence and inflation. Market-replacement cost of the technology is then derived by calculating the weighted average of adjusted costs and similarity values of similar R&D projects. A case of "Prevention method and system for the diffusion of mobile malicious code" is presented to illustrate the proposed approach.