• Journal of KIISE
• /
• v.43 no.6
• /
• pp.678-685
• /
• 2016

Named Entity Recognition and Classification (NERC) is a task for recognition and classification of named entities such as a person's name, location, and organization. There have been various studies carried out on Korean NERC, but they have some problems, for example lacking some features as compared with English NERC. In this paper, we propose a method that uses word embedding as features for Korean NERC. We generate a word vector using a Continuous-Bag-of-Word (CBOW) model from POS-tagged corpus, and a word cluster symbol using a K-means algorithm from a word vector. We use the word vector and word cluster symbol as word embedding features in Conditional Random Fields (CRFs). From the result of the experiment, performance improved 1.17%, 0.61% and 1.19% respectively for TV domain, Sports domain and IT domain over the baseline system. Showing better performance than other NERC systems, we demonstrate the effectiveness and efficiency of the proposed method.

• The Journal of Korea Robotics Society
• /
• v.12 no.1
• /
• pp.55-64
• /
• 2017

This paper proposes combination of a cognitive agent architecture named Soar (State, operator, and result) and ROS (Robot Operating System), which can be a basic framework for a robot agent to interact and cope with its environment more intelligently and appropriately. The proposed Soar-ROS human-robot interaction (HRI) agent understands a set of human's commands by voice recognition and chooses to properly react to the command according to the symbol detected by image recognition, implemented on a humanoid robot. The robotic agent is allowed to refuse to follow an inappropriate command like "go" after it has seen the symbol 'X' which represents that an abnormal or immoral situation has occurred. This simple but meaningful HRI task is successfully experimented on the proposed Soar-ROS platform with a small humanoid robot, which implies that extending the present hybrid platform to artificial moral agent is possible.

• 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.

• International Journal of Contents
• /
• v.10 no.4
• /
• pp.63-68
• /
• 2014

In this paper, we propose a smart system that can optically recognize a music score within a document and can play the music after recognition. Many historic handwritten documents have now been digitalized. Converting images of a music score within documents into digital files is particularly difficult and requires considerable resources because a music score consists of a 2D structure with both staff lines and symbols. The proposed system takes an input image using a mobile device equipped with a camera module, and the image is optimized via preprocessing. Binarization, music sheet correction, staff line recognition, vertical line detection, note recognition, and symbol recognition processing are then applied, and a music file is generated in an XML format. The Music XML file is recorded as digital information, and based on that file, we can modify the result, logically correct errors, and finally generate a MIDI file. Our system reduces misrecognition, and a wider range of music score can be recognized because we have implemented distortion correction and vertical line detection. We show that the proposed method is practical, and that is has potential for wide application through an experiment with a variety of music scores.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.3
• /
• pp.200-207
• /
• 2001

In this paper, we propose an HMM(Hidden Markov modeJ)-MLP(Multi-layer perceptron) hybrid model for recognizing legal words on the English bank check. We adopt an explicit segmentation-based word level architecture to implement an HMM engine with nonscaled and non-normalized symbol vectors. We also introduce an MLP for implicit segmentation-based word recognition. The final recognition model consists of a hybrid combination of the HMM and MLP with a new hybrid probability measure. The main contributions of this model are a novel design of the segmentation-based variable length HMMs and an efficient method of combining two heterogeneous recognition engines. ExperimenLs have been conducted using the legal word database of CENPARMI with encouraging results.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.10
• /
• pp.2409-2416
• /
• 2014

In this paper, an automatic recognition algorithm for linearly modulated signals like PSK, QAM under noncoherent asynchronous condition is proposed. Frequency, phase, and amplitude characteristics of digitally modulated signals are changed periodically. By using this characteristics, cyclic moments and higher order cumulants based features are utilized for the modulation recognition. Hierarchial decision tree method is used for high speed signal processing and totally 4 feature extraction parameters are used for modulation recognition. In the condition where the symbol number is 4,096, the recognition accuracy of the proposed algorithm is more than 95% at SNR 15dB. Also the proposed algorithm is effective to classify the signal which has carrier frequency and phase offset.

• Proceedings of the Korean Institute of Interior Design Conference
• /
• 2007.05a
• /
• pp.270-274
• /
• 2007

Clearing trace of symbol which was gone with a series of pre-modern architecture history since the modern architecture (pursuing true nature from tradition which is repeated and imitated unconsciously). That is, What is the course of deconstruction? In the early part of the 20th century, We still accept the necessity of decoration in spite of its existence at one time being threaten. This means, even though symbolism in architecture has relative importance by situation of Times, it plays an important role to add the past to current style through 'Symbol'. The history of Times, a carrier which reflects Present on New Futures, makes memory by gathering data but we can not amplify our historical imagination with only data. Data is a past memory and evidence but we can not substitute that for historical experience. And it is difficult for future generations who don't live through that history to change their historical recognition with recollecting memories. They have to draw history with data but it is very limited in itself. However, They can collect historical memory through symbol in architecture. In this study, We pay attention to the symbolism of a memorial hall architecture. So We'll analyze dichotomy concept of Barthes's signifiant and signifie, visual sign and course of symbolic meaning on basis of Daniel Libeskind's Jewish Museum.

• Journal of Information Processing Systems
• /
• v.11 no.3
• /
• pp.438-449
• /
• 2015

Recognition systems for scanned or printed music scores that have been implemented on personal computers have received attention from numerous scientists and have achieved significant results over many years. A modern trend with music scores being captured and played directly on mobile devices has become more interesting to researchers. The limitation of resources and the effects of illumination, distortion, and inclination on input images are still challenges to these recognition systems. In this paper, we introduce a novel approach for recognizing music scores captured by mobile cameras. To reduce the complexity, as well as the computational time of the system, we grouped all of the symbols extracted from music scores into ten main classes. We then applied each major class to SVM to classify the musical symbols separately. The experimental results showed that our proposed method could be applied to real time applications and that its performance is competitive with other methods.

• Journal of Korea Multimedia Society
• /
• v.15 no.10
• /
• pp.1196-1204
• /
• 2012

In this paper, a series of algorithms are proposed for controlling different kinds of multimedia contents and realizing interact between human and computer by using single input device. Human gesture recognition based on NUI is presented firstly in my paper. Since the image information we get it from camera is not sensitive for further processing, we transform it to YCbCr color space, and then morphological processing algorithm is used to delete unuseful noise. Boundary Energy and depth information is extracted for hand detection. After we receive the image of hand detection, PCA algorithm is used to recognize hand posture, difference image and moment method are used to detect hand centroid and extract trajectory of hand movement. 8 direction codes are defined for quantifying gesture trajectory, so the symbol value will be affirmed. Furthermore, HMM algorithm is used for hand gesture recognition based on the symbol value. According to series of methods we presented, we can control multimedia contents by using human gesture recognition. Through large numbers of experiments, the algorithms we presented have satisfying performance, hand detection rate is up to 94.25%, gesture recognition rate exceed 92.6%, hand posture recognition rate can achieve 85.86%, and face detection rate is up to 89.58%. According to these experiment results, we can control many kinds of multimedia contents on computer effectively, such as video player, MP3, e-book and so on.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.96-105
• /
• 1998

The aim of this paper is to propose the method of WDHMM(Weighted DHMM), using the MLP-VQ for the improvement of speaker-independent connect-digit recognition system. MLP neural-network output distribution shows a probability distribution that presents the degree of similarity between each pattern by the non-linear mapping among the input patterns and learning patterns. MLP-VQ is proposed in this paper. It generates codewords by using the output node index which can reach the highest level within MLP neural-network output distribution. Different from the old VQ, the true characteristics of this new MLP-VQ lie in that the degree of similarity between present input patterns and each learned class pattern could be reflected for the recognition model. WDHMM is also proposed. It can use the MLP neural-network output distribution as the way of weighing the symbol generation probability of DHMMs. This newly-suggested method could shorten the time of HMM parameter estimation and recognition. The reason is that it is not necessary to regard symbol generation probability as multi-dimensional normal distribution, as opposed to the old SCHMM. This could also improve the recognition ability by 14.7% higher than DHMM, owing to the increase of small caculation amount. Because it can reflect phone class relations to the recognition model. The result of my research shows that speaker-independent connected-digit recognition, using MLP-VQ and WDHMM, is 84.22%.