• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3230-3255
• /
• 2022

Causality mining in NLP is a significant area of interest, which benefits in many daily life applications, including decision making, business risk management, question answering, future event prediction, scenario generation, and information retrieval. Mining those causalities was a challenging and open problem for the prior non-statistical and statistical techniques using web sources that required hand-crafted linguistics patterns for feature engineering, which were subject to domain knowledge and required much human effort. Those studies overlooked implicit, ambiguous, and heterogeneous causality and focused on explicit causality mining. In contrast to statistical and non-statistical approaches, we present Bidirectional Encoder Representations from Transformers (BERT) integrated with Multi-level Feature Networks (MFN) for causality recognition, called BERT+MFN for causality recognition in noisy and informal web datasets without human-designed features. In our model, MFN consists of a three-column knowledge-oriented network (TC-KN), bi-LSTM, and Relation Network (RN) that mine causality information at the segment level. BERT captures semantic features at the word level. We perform experiments on Alternative Lexicalization (AltLexes) datasets. The experimental outcomes show that our model outperforms baseline causality and text mining techniques.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.348-353
• /
• 2022

최근 신경망 기반의 언어모델이 발전함에 따라 대부분의 검색 모델에서는 Bi-encoder를 기반으로한 Dense retrieval 모델에 대한 연구가 진행되고 있다. 특히 DPR은 BM25를 통해 정답 문서와 유사한 정보를 가진 하드 네거티브를 사용하여 대조학습을 통해 성능을 더욱 끌어올린다. 그러나 BM25로 검색된 하드 네거티브는 term-base의 유사도를 통해 뽑히기 때문에, 의미적으로 비슷한 내용을 갖는 하드 네거티브의 역할을 제대로 수행하지 못하고 대조학습의 효율성을 낮출 가능성이 있다. 따라서 DRP의 대조학습에서 하드 네거티브의 역할을 본질적으로 수행할 수 있는 문서를 샘플링 하는 방법을 제시하고, 이때 얻은 하드 네거티브의 집합을 주기적으로 업데이트 하여 효과적으로 대조학습을 진행하는 방법을 제안한다. 지식 기반 대화 데이터셋인 MultiDoc2Dial을 통해 평가를 수행하였으며, 실험 결과 기존 방식보다 더 높은 성능을 나타낸다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.617-621
• /
• 2021

의미역 결정은 문장 속에서 서술어와 그 논항의 관계를 파악하며, '누가, 무엇을, 어떻게, 왜' 등과 같은 의미역 관계를 찾아내는 자연어 처리 기법이다. 최근 수행되고 있는 의미역 결정 연구는 주로 말뭉치를 활용하여 딥러닝 학습을 하는 방식으로 연구가 이루어지고 있다. 최근 구글에서 개발한 사전 훈련된 Bidirectional Encoder Representations from Transformers (BERT) 모델이 다양한 자연어 처리 분야에서 상당히 높은 성능을 보이고 있다. 본 논문에서는 한국어 의미역 결정 성능 향상을 위해 한국어의 언어적 특징을 고려하며 사전 학습된 SNU KR-BERT를 사용하면서 한국어 의미역 결정 모델의 성능을 살펴보였다. 또한, 본 논문에서는 BERT 모델에서 과연 어떤 히든 레이어(hidden layer)에서 한국어 의미역 결정을 더 잘 수행하는지 알아보고자 하였다. 실험 결과 마지막 히든 레이어 임베딩을 활용하였을 때, 언어 모델의 성능은 66.4% 였다. 히든 레이어 별 언어 모델 성능을 비교한 결과, 마지막 4개의 히든 레이어를 이었을 때(concatenated), 언어 모델의 성능은 67.9% 이였으며, 11번째 히든 레이어를 사용했을 때는 68.1% 이였다. 즉, 마지막 히든 레이어를 선택했을 때보다 더 성능이 좋았다는 것을 알 수 있었다. 하지만 각 언어 모델 별 히트맵을 그려보았을 때는 마지막 히든 레이어 임베딩을 활용한 언어 모델이 더 정확히 의미역 판단을 한다는 것을 알 수 있었다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.11
• /
• pp.441-448
• /
• 2019

Previous researches on automatic spacing of Korean sentences has been researched to correct spacing errors by using n-gram based statistical techniques or morpheme analyzer to insert blanks in the word boundary. In this paper, we propose an end-to-end automatic word spacing by using deep neural network. Automatic word spacing problem could be defined as a tag classification problem in unit of syllable other than word. For contextual representation between syllables, Bi-LSTM encodes the dependency relationship between syllables into a fixed-length vector of continuous vector space using forward and backward LSTM cell. In order to conduct automatic word spacing of Korean sentences, after a fixed-length contextual vector by Bi-LSTM is classified into auto-spacing tag(B or I), the blank is inserted in the front of B tag. For tag classification method, we compose three types of classification neural networks. One is feedforward neural network, another is neural network language model and the other is linear-chain CRF. To compare our models, we measure the performance of automatic word spacing depending on the three of classification networks. linear-chain CRF of them used as classification neural network shows better performance than other models. We used KCC150 corpus as a training and testing data.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.6
• /
• pp.736-745
• /
• 2002

In network delivery of compressed video, packets may be lost if the channel is unreliable like Internet. Such losses tend to of cur in burst like continuous bit-stream error. In this paper, we propose an effective error-concealment approach to which an error resilient video encoding approach is applied against burst errors and which reduces a complexity of error concealment at the decoder using data hiding. To improve the performance of error concealment, a temporal and spatial error resilient video encoding approach at encoder is developed to be robust against burst errors. For spatial area of error concealment, block shuffling scheme is introduced to isolate erroneous blocks caused by packet losses. For temporal area of error concealment, we embed parity bits in content data for motion vectors between intra frames or continuous inter frames and recovery loss packet with it at decoder after transmission While error concealment is performed on error blocks of video data at decoder, it is computationally costly to interpolate error video block using neighboring information. So, in this paper, a set of feature are extracted at the encoder and embedded imperceptibly into the original media. If some part of the media data is damaged during transmission, the embedded features can be extracted and used for recovery of lost data with bi-direction interpolation. The use of data hiding leads to reduced complexity at the decoder. Experimental results suggest that our approach can achieve a reasonable quality for packet loss up to 30% over a wide range of video materials.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.4
• /
• pp.593-603
• /
• 2020

The advent of malicious code has increased exponentially due to the spread of malicious code generation tools in accordance with the development of the network, but there is a limit to the response through existing malicious code detection methods. According to this situation, a machine learning-based malicious code detection method is evolving, and in this paper, the feature of data is extracted from the PE header for machine-learning-based malicious code detection, and then it is used to automate the malware through autoencoder. Research on how to extract the indicated features and feature importance. In this paper, 549 features composed of information such as DLL/API that can be identified from PE files that are commonly used in malware analysis are extracted, and autoencoder is used through the extracted features to improve the performance of malware detection in machine learning. It was proved to be successful in providing excellent accuracy and reducing the processing time by 2 times by effectively extracting the features of the data by compressively storing the data. The test results have been shown to be useful for classifying malware groups, and in the future, a classifier such as SVM will be introduced to continue research for more accurate malware detection.