• Journal of KIISE:Software and Applications
• /
• v.29 no.4
• /
• pp.258-269
• /
• 2002

There have been many studies of automatic term recognition (ATR) and they have achieved good results. However, there are scopes to improve the performance of extracting terms still further by using the additional technical dictionaries. This paper focuses on the method for extracting terms using the hierarchy among technical dictionaries. Moreover, a statistical method based on frequencies, foreign words, and nested relations assists extracting terms which do not appear in dictionaries. Our method produces relatively good results for this task.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.4E
• /
• pp.112-119
• /
• 2004

This paper presents a method for the automatic recognition of pitch accents with no prior knowledge about the phonetic content of the signal (no knowledge of word or phoneme boundaries or of phoneme labels). The recognition algorithm used in this paper is a time-delay recurrent neural network (TDRNN). A TDRNN is a neural network classier with two different representations of dynamic context: delayed input nodes allow the representation of an explicit trajectory F0(t), while recurrent nodes provide long-term context information that can be used to normalize the input F0 trajectory. Performance of the TDRNN is compared to the performance of a MLP (multi-layer perceptron) and an HMM (Hidden Markov Model) on the same task. The TDRNN shows the correct recognition of $91.9{\%}\;of\;pitch\;events\;and\;91.0{\%}$ of pitch non-events, for an average accuracy of $91.5{\%}$ over both pitch events and non-events. The MLP with contextual input exhibits $85.8{\%},\;85.5{\%},\;and\;85.6{\%}$ recognition accuracy respectively, while the HMM shows the correct recognition of $36.8{\%}\;of\;pitch\;events\;and\;87.3{\%}$ of pitch non-events, for an average accuracy of $62.2{\%}$ over both pitch events and non-events. These results suggest that the TDRNN architecture is useful for the automatic recognition of pitch accents.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2000.05a
• /
• pp.131-136
• /
• 2000

기존의 통계에 기반한 용어 자동 추출 기법(Automatic Term Recognition)은 비교적 좋은 성능의 결과를 보여왔다. 하지만 전문용어 사전 등의 정보를 이용하여 성능의 향상을 이룰 수 있는 여지는 여전히 남아있다. 본 논문에서는 이러한 근거에 기반하여 전문용어간의 계층 정보를 전문용어 사전을 통하여 구축하고 이를 이용하여 전문용어를 추출하는 방법을 제안하고자 한다. 본 논문이 제안하는 기법은 기존의 방법에 비해 좋은 성능을 나타내었다.

• ETRI Journal
• /
• v.42 no.5
• /
• pp.761-772
• /
• 2020

This paper presents an automatic proficiency assessment method for a non-native Korean read utterance using bidirectional long short-term memory (BLSTM)-based acoustic models (AMs) and speech data augmentation techniques. Specifically, the proposed method considers two scenarios, with and without prompted text. The proposed method with the prompted text performs (a) a speech feature extraction step, (b) a forced-alignment step using a native AM and non-native AM, and (c) a linear regression-based proficiency scoring step for the five proficiency scores. Meanwhile, the proposed method without the prompted text additionally performs Korean speech recognition and a subword un-segmentation for the missing text. The experimental results indicate that the proposed method with prompted text improves the performance for all scores when compared to a method employing conventional AMs. In addition, the proposed method without the prompted text has a fluency score performance comparable to that of the method with prompted text.

• Proceedings of the CALSEC Conference
• /
• 2005.03a
• /
• pp.55-61
• /
• 2005

We will introduce a new web-based knowledge management system in progress, in which XML-based web information extraction and our structuring knowledge technologies are combined using ontology-based natural language processing. Our aim is to provide efficient access to heterogeneous information on the web, enabling users to use a wide range of textual and non textual resources, such as newspapers and databases, effortlessly to accelerate knowledge acquisition from such knowledge sources. In order to achieve the efficient knowledge management, we propose at first an XML-based Web information extraction which contains a sophisticated control language to extract data from Web pages. With using standard XML Technologies in the system, our approach can make extracting information easy because of a) detaching rules from processing, b) restricting target for processing, c) Interactive operations for developing extracting rules. Then we propose a structuring knowledge system which includes, 1) automatic term recognition, 2) domain oriented automatic term clustering, 3) similarity-based document retrieval, 4) real-time document clustering, and 5) visualization. The system supports integrating different types of databases (textual and non textual) and retrieving different types of information simultaneously. Through further explanation to the specification and the implementation technique of the system, we will demonstrate how the system can accelerate knowledge acquisition on the Web even for novice users of the field.

• Journal of KIISE:Software and Applications
• /
• v.33 no.8
• /
• pp.718-729
• /
• 2006

There has been increasing interest in automatic term recognition (ATR), which recognizes technical terms for given domain specific texts. ATR is composed of 'term extraction', which extracts candidates of technical terms and 'term selection' which decides whether terms in a term list derived from 'term extraction' are technical terms or not. 'term selection' is a process to rank a term list depending on features of technical term and to find the boundary between technical term and general term. The previous works just use statistical features of terms for 'term selection'. However, there are limitations on effectively selecting technical terms among a term list using the statistical feature. The objective of this paper is to find effective features for 'term selection' by considering various aspects of technical terms. In order to solve the ranking problem, we derive various features of technical terms and combine the features using machine-learning algorithms. For solving the boundary finding problem, we define it as a binary classification problem which classifies a term in a term list into technical term and general term. Experiments show that our method records 78-86% precision and 87%-90% recall in boundary finding, and 89%-92% 11-point precision in ranking. Moreover, our method shows higher performance than the previous work's about 26% in maximum.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1079-1089
• /
• 2023

Stroke is one of the leading causes of long-term disability worldwide, placing huge burdens on individuals and society. Further, automatic human activity recognition is a challenging task that is vital to the future of healthcare and physical therapy. Using a baseline long short-term memory recurrent neural network, this study provides a novel dataset of stretching, upward stretching, flinging motions, hand-to-mouth movements, swiping gestures, and pouring motions for improved model training and testing of stroke-affected patients. A MATLAB application is used to output textual and audible prediction results. A wearable sensor with a triaxial accelerometer is used to collect preprocessed real-time data. The model is trained with features extracted from the actual patient to recognize new actions, and the recognition accuracy provided by multiple datasets is compared based on the same baseline model. When training and testing using the new dataset, the baseline model shows recognition accuracy that is 11% higher than the Activity Daily Living dataset, 22% higher than the Activity Recognition Single Chest-Mounted Accelerometer dataset, and 10% higher than another real-world dataset.

• Journal of Korea Multimedia Society
• /
• v.19 no.3
• /
• pp.612-619
• /
• 2016

Applying evolved IT technology to increase the satisfaction of the consumer is a typical feature of the rapidly transformed service industry. Smart devices have taken a place as the mainstream of major media, moreover, IT service strategies utilizing smart devices have been constantly developed. Location recognition method of users has been proposed as one of the significant features to the IT service industry. The long-term care home services can be one example of user location recognition methods that real time computerization of service record utilize an attached tag to home of pensioners with RFID reader or an NFC function of mobile when a home carer provides service. And, the Homecare Electronics Management System (HEMS) and the various location recognition methods will be discussed to improve effectiveness of services. In this paper, we propose a home applicable electronic management system which insurer, home care service facilities, home cares, pensioners and guardians which enables to simultaneously check service records based on the improved system by use of iBeacon.

• Journal of Information Processing Systems
• /
• v.18 no.4
• /
• pp.480-488
• /
• 2022

Images of tobacco retail licenses have complex unstructured characteristics, which is an urgent technical problem in the robot process automation of tobacco marketing. In this paper, a novel recognition approach using a double attention mechanism is presented to realize the automatic recognition and information extraction from such images. First, we utilized a DenseNet network to extract the license information from the input tobacco retail license data. Second, bi-directional long short-term memory was used for coding and decoding using a continuous decoder integrating dual attention to realize the recognition and information extraction of tobacco retail license images without segmentation. Finally, several performance experiments were conducted using a largescale dataset of tobacco retail licenses. The experimental results show that the proposed approach achieves a correction accuracy of 98.36% on the ZY-LQ dataset, outperforming most existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.10
• /
• pp.1275-1281
• /
• 2019

The automatic modulation recognition of a radio signal is a major task of an intelligent receiver, with various civilian and military applications. In this paper, we propose a method to recognize the modulation of radio signals in wireless communication based on the deep neural network. We classify the modulation pattern of radio signal by using the LSTM model, which can catch the long-term pattern for the sequential data as the input data of the deep neural network. The amplitude and phase of the modulated signal, the in-phase carrier, and the quadrature-phase carrier are used as input data in the LSTM model. In order to verify the performance of the proposed learning method, we use a large dataset for training and test, including the ten types of modulation signal under various signal-to-noise ratios.