• The KIPS Transactions:PartB
• /
• v.12B no.2 s.98
• /
• pp.151-156
• /
• 2005

In this paper, we propose feature based character watermarking methods based on geometical features specific to characters of text in document image. The proposed methods can satisfy both data capacity and robustness simultaneously while none of the conventional methods can. According to the characteristics of characters, watermark can be embed or detected through changes of connectivity of the characters, differences of characteristics of edge pixels or changes of area of holes. Experimental results show that our identification techniques are very robust to distortion and have high data capacity.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.105-108
• /
• 2010

쌍방향으로 질문과 답변을 하는 커뮤니티 기반의 지식검색서비스에서는 질의를 통해 원하는 답변을 얻을 수 있지만, 수많은 사용자들이 참여함에 따라 방대한 문서 속에서 검증된 문서를 찾아내는 것은 점점 더 어려워지고 있다. 지식검색서비스에서 기존 연구는 사용자들이 생성한 데이터 즉 추천수, 조회수 등의 비텍스트 정보를 이용하거나 답변의 길이, 자료첨부, 연결어 등의 텍스트 정보 이용하여 전문가를 식별하거나 문서의 품질을 평가하고, 이를 검색에 반영하여 검색성능을 향상시키는 데 활용했다. 그러나 비텍스트 정보는 질의/응답의 초기에 사용자들에 의해 충분한 정보를 확보할 수 없는 단점이 제기 되며, 텍스트 정보는 전체의 문서를 답변의 길이, 자료 첨부등과 같은 일부요인으로 판단해야하기 때문에 품질평가의 한계가 있다고 볼 수 있겠다. 본 논문에서는 이러한 비텍스트 정보와 텍스트 정보의 문제점을 개선하기 위한 품질평가 알고리즘을 제안한다. 제안된 알고리즘을 통한 품질지수는 텍스트/비텍스트 정보와 소셜 네트워크 사용자 중앙성을 고려하여 질문에 적합하고 신뢰성 있는 답변을 랭킹화 함으로써 지식검색문서를 분별하는 지표가 되며, 이는 지식검색서비스의 성능향상에 기여를 할 수 있을 것으로 기대된다.

• Computers and Concrete
• /
• v.16 no.2
• /
• pp.287-309
• /
• 2015

FormWorks-Plus is a generalized public domain user-friendly preprocessor developed to facilitate the process of creating finite element models for structural analysis programs. The lack of a graphical user interface in most academic analysis programs forces users to input the structural model information into the standard text files, which is a time-consuming and error-prone process. FormWorks-Plus enables engineers to conveniently set up the finite element model in a graphical environment, eliminating the problems associated with conventional input text files and improving the user's perception of the application. In this paper, a brief overview of the FormWorks-Plus structure is presented, followed by a detailed explanation of the main features of the program. In addition, demonstration is made of the application of FormWorks-Plus in combination with VecTor programs, advanced nonlinear analysis tools for reinforced concrete structures. Finally, aspects relating to the modelling and analysis of three case studies are discussed: a reinforced concrete beam-column joint, a steel-concrete composite shear wall, and a SFRC shear panel. The unique mixed-type frame-membrane modelling procedure implemented in FormWorks-Plus can address the limitations associated with most frame type analyses.

• Genomics & Informatics
• /
• v.17 no.2
• /
• pp.15.1-15.7
• /
• 2019

Automatically detecting mentions of pharmaceutical drugs and chemical substances is key for the subsequent extraction of relations of chemicals with other biomedical entities such as genes, proteins, diseases, adverse reactions or symptoms. The identification of drug mentions is also a prior step for complex event types such as drug dosage recognition, duration of medical treatments or drug repurposing. Formally, this task is known as named entity recognition (NER), meaning automatically identifying mentions of predefined entities of interest in running text. In the domain of medical texts, for chemical entity recognition (CER), techniques based on hand-crafted rules and graph-based models can provide adequate performance. In the recent years, the field of natural language processing has mainly pivoted to deep learning and state-of-the-art results for most tasks involving natural language are usually obtained with artificial neural networks. Competitive resources for drug name recognition in English medical texts are already available and heavily used, while for other languages such as Spanish these tools, although clearly needed were missing. In this work, we adapt an existing neural NER system, NeuroNER, to the particular domain of Spanish clinical case texts, and extend the neural network to be able to take into account additional features apart from the plain text. NeuroNER can be considered a competitive baseline system for Spanish drug and CER promoted by the Spanish national plan for the advancement of language technologies (Plan TL).

• International journal of advanced smart convergence
• /
• v.10 no.1
• /
• pp.97-104
• /
• 2021

Securities and investment services had the most data per company on the average, and used the most data. Investors are increasingly demanding to invest through their own analysis methods. Therefore, securities and investment companies provide stock data to investors through open API. The data received using the open API is in text format. Python is effective and convenient for requesting and receiving text data. We investigate there are 22 major securities and investment companies in Korea and only 6 companies. Only Daishin Securities Co. supports Python officially. We compare how to receive stock data through open API using Python, and Python programming features. The open APIs for the study are Daishin Securities Co. and eBest Investment & Securities Co. Comparing the two APIs for receiving the current stock data, we find the main two differences are the login method and the method of sending and receiving data. As for the login method, CYBOS plus has login information, but xingAPI does not have. As for the method of sending and receiving data, Cybos Plus sends and receives data by calling the request method, and the reply method. xingAPI sends and receives data in the form of an event. Therefore, the number of xingAPI codes is more than that of CYBOS plus. And we find that CYBOS plus executes a loop statement by lists and tuple, dictionary, and CYBOS plus supports the basic commands provided by Python.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.231-244
• /
• 2022

With the improvement of people's living standards and the development of tourism, tourists have greater freedom in choosing destinations. Therefore, as an indicator of satisfaction with scenic spots, tourist comments are becoming increasingly prominent. This paper aims to compare and analyze the landscape image of the Five Great Mountains in China and provide specific strategies for its development. The online reviews of tourists on the Online Travel Agency (OTA) website about the Five Great Mountains from 2015 to 2018 are collected as research samples. The text analysis method and R language are used to analyze the content of the tourist reviews, while the high-frequency words in the word cloud are used for visual display. In addition, the entropy weight method is used to determine the index weight and tourist satisfaction is evaluated to understand the weaknesses of those scenic spots. The results of the study show that firstly, the tourist satisfaction with the Five Great Mountains is basically consistent with its popularity. Secondly, through weight analysis, tourists pay special attention to the landscape features and environmental health of the scenic area, so that relevant departments should focus on building the landscape characteristics and improving the environmental health of the scenic area. At the same time, the accommodation and service management of the scenic spot cannot be ignored. Finally, according to the analysis results, suggestions are made on how to improve the tourist satisfaction with the Five Great Mountains.

• Journal of Information Processing Systems
• /
• v.18 no.4
• /
• pp.535-548
• /
• 2022

Using a rich resource language to classify sentiments in a language with few resources is a popular subject of research in natural language processing. Burmese is a low-resource language. In light of the scarcity of labeled training data for sentiment classification in Burmese, in this study, we propose a method of transfer learning for sentiment analysis of a language that uses the feature transfer technique on sentiments in English. This method generates a cross-language word-embedding representation of Burmese vocabulary to map Burmese text to the semantic space of English text. A model to classify sentiments in English is then pre-trained using a convolutional neural network and an attention mechanism, where the network shares the model for sentiment analysis of English. The parameters of the network layer are used to learn the cross-language features of the sentiments, which are then transferred to the model to classify sentiments in Burmese. Finally, the model was tuned using the labeled Burmese data. The results of the experiments show that the proposed method can significantly improve the classification of sentiments in Burmese compared to a model trained using only a Burmese corpus.

• International Journal of Computer Science & Network Security
• /
• v.23 no.4
• /
• pp.15-24
• /
• 2023

One of the objectives and aspirations of scientists and engineers ever since the development of computers has been to interact naturally with machines. Hence features of artificial intelligence (AI) like natural language processing and natural language generation were developed. The field of AI that is thought to be expanding the fastest is interactive conversational systems. Numerous businesses have created various Virtual Personal Assistants (VPAs) using these technologies, including Apple's Siri, Amazon's Alexa, and Google Assistant, among others. Even though many chatbots have been introduced through the years to diagnose or treat psychological disorders, we are yet to have a user-friendly chatbot available. A smart generative cognitive behavioral therapy with spoken dialogue systems support was then developed using a model Persona Perception (P2) bot with Generative Pre-trained Transformer-2 (GPT-2). The model was then implemented using modern technologies in VPAs like voice recognition, Natural Language Understanding (NLU), and text-to-speech. This system is a magnificent device to help with voice-based systems because it can have therapeutic discussions with the users utilizing text and vocal interactive user experience.

• Journal of The Korean Association For Science Education
• /
• v.40 no.1
• /
• pp.61-75
• /
• 2020

In this study, external representation presented in two units, 'Property of Gas' and 'Changes of States of Matter,' in seventh grade of 2015 revised science curriculum, were analyzed to suggest educational implications. External representations presented in five science textbooks were analyzed according to the six criteria, which were 'type of representation,' 'interpretation of surface features,' 'relatedness to text,' 'existence and properties of a caption,' 'degree of correlation between representations comprising a multiple one,' and 'function of representation.' The characteristics of typical representations related to each achievement standard of two units were also analyzed. The results were as follows: The macro representations for 'type of representation', and explicit representations for 'interpretation of surface features' showed highest frequency. For 'relatedness to text' criteria, 'completely related and linked' and 'completely related and unlinked' representations showed the highest frequency. It means that most representations were properly related with the text. There were appropriate captions for most representations. The degree of correlation between representations comprising a multiple one was largely sufficiently linked with regards to the criteria 'degree of correlation between representations comprising a multiple one'. The complete representations for 'function of representation' showed the highest frequency in the aggregate, however incomplete representations showed more frequencies in the inquiry parts. The typical representations for each achievement standard differed in terms of the type, contained information, used symbols and so on. The educational implications with the use of representations presented in seventh grade textbook were discussed.

• Journal of Internet Computing and Services
• /
• v.21 no.1
• /
• pp.45-55
• /
• 2020

Due to the development of information and communication technology, the number of new / variant malicious codes is increasing rapidly every year, and various types of malicious codes are spreading due to the development of Internet of things and cloud computing technology. In this paper, we propose a malware analysis method based on string information that can be used regardless of operating system environment and represents library call information related to malicious behavior. Attackers can easily create malware using existing code or by using automated authoring tools, and the generated malware operates in a similar way to existing malware. Since most of the strings that can be extracted from malicious code are composed of information closely related to malicious behavior, it is processed by weighting data features using text mining based method to extract them as effective features for malware analysis. Based on the processed data, a model is constructed using various machine learning algorithms to perform experiments on detection of malicious status and classification of malicious groups. Data has been compared and verified against all files used on Windows and Linux operating systems. The accuracy of malicious detection is about 93.5%, the accuracy of group classification is about 90%. The proposed technique has a wide range of applications because it is relatively simple, fast, and operating system independent as a single model because it is not necessary to build a model for each group when classifying malicious groups. In addition, since the string information is extracted through static analysis, it can be processed faster than the analysis method that directly executes the code.