• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.870-878
• /
• 2019

In this paper, we propose an Attention-based BLSTM for predicting the video compression standard of a video. Recently, in NLP, many researches have been studied to predict the next word of sentences, classify and translate sentences by their semantics using the structure of RNN, and they were commercialized as chatbots, AI speakers and translator applications, etc. LSTM is designed to solve the gradient vanishing problem in RNN, and is used in NLP. The proposed algorithm makes video compression standard prediction possible by applying BLSTM and Attention algorithm which focuses on the most important word in a sentence to a bitstream of a video, not an sentence of a natural language.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.4
• /
• pp.163-170
• /
• 2019

Korea language has postpositions such as eul, reul, yi, ga, wa, and gwa, which are attached to nouns and add meaning to the sentence. When foreign notations or abbreviations are included in sentences, the appropriate postposition for the pronunciation of the foreign words may not be used. Sometimes, for natural expression of the sentence, two postpositions are used with one in parentheses as in "eul(reul)" so that both postpositions can be acceptable. This study finds examples of using unnatural postpositions when foreign words are included in Korean sentences and proposes a method for using natural postpositions by learning the final consonant pronunciation of nouns. The proposed method uses a recurrent neural network model to naturally express postpositions connected to foreign words. Furthermore, the proposed method is proven by learning and testing with the proposed method. It will be useful for composing perfect sentences for machine translation by using natural postpositions for English abbreviations or new foreign words included in Korean sentences in the future.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.281-291
• /
• 2021

The Islamic jurisprudential legal system represents an essential component of the Islamic religion, that governs many aspects of Muslims' daily lives. This creates many questions that require interpretations by qualified specialists, or Muftis according to the main sources of legislation in Islam. The Islamic jurisprudence is usually classified into branches, according to which the questions can be categorized and classified. Such categorization has many applications in automated question-answering systems, and in manual systems in routing the questions to a specialized Mufti to answer specific topics. In this work we tackle the problem of automatic categorisation of Islamic jurisprudential legal questions using deep learning techniques. In this paper, we build a hierarchical deep learning model that first extracts the question text features at two levels: word and sentence representation, followed by a text classifier that acts upon the question representation. To evaluate our model, we build and release the largest publicly available dataset of Islamic questions and answers, along with their topics, for 52 topic categories. We evaluate different state-of-the art deep learning models, both for word and sentence embeddings, comparing recurrent and transformer-based techniques, and performing extensive ablation studies to show the effect of each model choice. Our hierarchical model is based on pre-trained models, taking advantage of the recent advancement of transfer learning techniques, focused on Arabic language.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.1
• /
• pp.49-62
• /
• 2022

Online customer review data can be easily collected on the Internet and also they describe sentimental evaluation of a product in different aspects. Previous sentiment analysis studies evaluate the degree of sentiment with review data, which may have multiple sentences describing different product aspects. Since different aspects of a product can be described in a sentence, the proposed method suggested analyzing a sentence to build a pair of a product aspect terms and sentimental terms. Bidirectional LSTM and CRF algorithms were used in this paper. A pair of aspect terms and sentimental terms are evaluated by pre-defined evaluation rules. The paper suggested using the result of evaulation as inputs of QFD, so that the quantified customer voices effect on the requirements of a new product. Online reviews for a hair dryer were used as an example showing that the proposed approach can derive reasonable sentiment analysis results.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.11
• /
• pp.521-528
• /
• 2021

Existing image analysis systems in use in the military field are carried out by readers analyzing and identifying images themselves, writing and disseminating related content, and in this process, repetitive tasks are frequent, resulting in workload. In this paper, to solve the previous problem, we proposed an algorithm that can operate the Seq2Seq model on a word basis, which operates on a sentence basis, and applied the Attention technique to improve accuracy. In addition, by applying the Beam Search technique, we would like to recommend various current identification sentences based on the past identification contents of a specific area. It was confirmed through experiments that the Beam Search technique recommends sentences more effectively than the existing greedy Search technique, and confirmed that the accuracy of recommendation increases when the size of Beam is large.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.1
• /
• pp.61-69
• /
• 2009

This paper presents an improved method measuring inter-utterance similarity in an example-based dialogue system, which searches the most similar utterance in a dialogue database to generate a response to a given user utterance. Unlike general inter-sentence similarity measures, the inter-utterance similarity measure for example-based dialogue system should consider not only word distribution but also various linguistic features, such as affirmation/negation, tense, modality, sentence type, which affects the natural conversation. However, previous approaches do not sufficiently reflect these features. This paper proposes a new utterance similarity measure by analyzing and reflecting various linguistic features to improve performance in accuracy. Also, by considering substitutability of the features, the proposed method can utilize limited number of examples. Experimental results show that the proposed method achieves 10%p improvement in accuracy compared to the previous method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.1404-1407
• /
• 2007

본 논문에서는 객관식 문항을 난이도에 따라 자동으로 생성하는 방법을 고안하여, 학습자 수준에 적합하도록 다양하고 동적인 형태로 문항 제시를 할 수 있는 시스템을 제안하였다. 이를 위해서는 주어진 문장에서 형태소 분석을 통해 키워드를 추출하고, 각 키워드에 대하여 워드넷의 계층적 특성에 따라 의미가 유사한 후보 단어를 제시한다. 의미 유사 후보 단어를 제시할 때, 워드넷에서의 어휘간 유사도 측정 방법을 사용함으로써 생성된 문항의 난이도를 사용자가 원하는 수준으로 조정할 수 있도록 하였다. 단어의 의미 유사도는 동의어를 의미하는 수준 0에서 거의 유사도를 찾을 수 없는 수준 9 까지 다양하게 제시할 수 있으며, 이를 조절함으로써 문항의 전체 난이도를 조절할 수 있다. 후보 어휘들의 의미 유사도 측정을 위해서, 본 논문에서는 두 가지 방법을 사용하여 구현하였다. 첫째는 단순히 두 어휘의 워드넷 상에서의 거리만을 고려한 것이고 둘째는 두 어휘가 워드넷에서 차지하는 비중까지 추가적으로 고려한 것이다. 이러한 방법을 통하여 실제 출제자가 기존에 출제된 문제를 토대로 보다 다양한 내용과 난이도를 가진 문제 또는 문항을 보다 쉽게 출제하게 함으로써 출제에 소요되는 비용을 줄일 수 있었다.

• Journal of Information Processing Systems
• /
• v.15 no.6
• /
• pp.1365-1377
• /
• 2019

Summarizing the differences in Chinese-Vietnamese bilingual news plays an important supporting role in the comparative analysis of news views between China and Vietnam. Aiming at cross-language problems in the analysis of the differences between Chinese and Vietnamese bilingual news, we propose a new method of summarizing the differences based on an undirected graph model. The method extracts elements to represent the sentences, and builds a bridge between different languages based on Wikipedia's multilingual concept description page. Firstly, we calculate the similarity between Chinese and Vietnamese news sentences, and filter the bilingual sentences accordingly. Then we use the filtered sentences as nodes and the similarity grade as the weight of the edge to construct an undirected graph model. Finally, combining the random walk algorithm, the weight of the node is calculated according to the weight of the edge, and sentences with highest weight can be extracted as the difference summary. The experiment results show that our proposed approach achieved the highest score of 0.1837 on the annotated test set, which outperforms the state-of-the-art summarization models.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.3
• /
• pp.688-702
• /
• 1998

In this paper, we propose sign language morpheme generation rule corresponding to morpheme analysis for each part of speech. Korean natural sign language has extremely limited vocabulary, and the number of grammatical components eing currently used are limited, too. In this paper, therefore, we define natural sign language grammar corresponding to Korean language grammar in order to translate natural Korean language sentences to the corresponding sign language. Each phrase should define sign language morpheme generation grammar which is different from Korean language analysis grammar. Then, this grammar is applied to morpheme analysis/combination rule and sentence structure analysis rule. It will make us generate most natural sign language by definition of this grammar.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.12a
• /
• pp.181-184
• /
• 2000

The new method of feature extraction is proposed, considering the statistic feature of human voice, unlike the conventional methods of voice extraction. PCA(principal Component Analysis) is applied to this new method. PCA removes the repeating of data after finding the axis direction which has the greatest variance in input dimension. Then the new method is applied to real voice recognition to assess performance. When results of the number recognition in this paper and the conventional Mel-Cepstrum of voice feature parameter are compared, there is 0.5% difference of recognition rate. Better recognition rate is expected than word or sentence recognition in that less convergence time than the conventional method in extracting voice feature. Also, better recognition tate is expected when the optimum vector is used by statistic feature of data.