• The Journal of Korea Robotics Society
• /
• v.5 no.3
• /
• pp.240-250
• /
• 2010

A URC, which is a Ubiquitous Robot Companion, provides services to users in ubiquitous computing environments and has advantage of simplifying robot's hardware and software by distributing the complicated functionality of robots to other system. In this paper, we propose SOWL, which is a software architecture for URC robots and a mixed word of SOMAR and CAWL. SOWL keeps the advantages of URC and it also has the loosely-coupled characteristics. Moreover it makes it easy to develop of URC robot software. The proposed architecture is composed of 4 layers: device software, robot software, robot application, and end user layer. Developers of the each layer is able to build software suitable for their requirements by combining software modules in the lower layer. SOWL consists of SOMAR and CAWL engine. SOMAR, which is a middleware for the execution of device software and robot software, is based on service-oriented architecture(SOA) for robot software. CAWL engine is a system to process CAWL which is a context-aware workflow language. SOWL is able to provide a layered architecture for the execution of a robot software. It also makes it possible for developers of the each layer to build module-based robot software.

• ETRI Journal
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2019

In this paper, we propose a rank-weighted reconstruction feature to improve the robustness of a feed-forward deep neural network (FFDNN)-based acoustic model. In the FFDNN-based acoustic model, an input feature is constructed by vectorizing a submatrix that is created by slicing the feature vectors of frames within a context window. In this type of feature construction, the appropriate context window size is important because it determines the amount of trivial or discriminative information, such as redundancy, or temporal context of the input features. However, we ascertained whether a single parameter is sufficiently able to control the quantity of information. Therefore, we investigated the input feature construction from the perspectives of rank and nullity, and proposed a rank-weighted reconstruction feature herein, that allows for the retention of speech information components and the reduction in trivial components. The proposed method was evaluated in the TIMIT phone recognition and Wall Street Journal (WSJ) domains. The proposed method reduced the phone error rate of the TIMIT domain from 18.4% to 18.0%, and the word error rate of the WSJ domain from 4.70% to 4.43%.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.10
• /
• pp.781-785
• /
• 2009

We can constantly gather personal life log from developed mobile device. However, gathered personal life log in mobile environment have a large amount log and uncertainty such as uncertainty of mobile environment, limited capacity and battery of mobile device. Tagging task using a landmark such as a key word should be required to overcome the above problem and to manage personal life log. In this paper, we propose new tagging method and a life log tagging interface using high level context recognizer based on probability. The new tagging method extract high level context such as landmark of life log using recognizer which is modeled from bayesian network and recommend recognized high level context to user using tagging interface. Finally user can directly do tagging task to life log. This task is a special feature in our process. As the result of experiments in task support level which include usability, level of a goal, function and leading, we achieved a feeling of satisfaction of 81%.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.4
• /
• pp.621-625
• /
• 2018

This paper proposes a method for bridge damage factor recognition from inspection reports using deep learning. Bridge inspection reports contains inspection results including identified damages and causal analysis results. However, collecting such information from inspection reports manually is limited due to their considerable amount. Therefore, this paper proposes a model for recognizing bridge damage factor from inspection reports applying Named Entity Recognition (NER) using deep learning. Named Entity Recognition, Word Embedding, Recurrent Neural Network, one of deep learning methods, were applied to construct the proposed model. Experimental results showed that the proposed model has abilities to 1) recognize damage and damage factor included in a training data, 2) distinguish a specific word as a damage or a damage factor, depending on its context, and 3) recognize new damage words not included in a training data.

• The KIPS Transactions:PartB
• /
• v.15B no.2
• /
• pp.171-178
• /
• 2008

Foreign word transliterations are not consistently used in documents, which hinders retrieving some important relevant documents in exact term matching information retrieval systems. In this paper, a meta search method is proposed, which expands and refines relevant variant queries from an original input foreign word transliteration query to retrieve the more relevant documents. The method firstly expands a transliteration query to the variants using a statistical method. Secondly the method selects the valid variants: it queries each variant to the retrieval systems beforehand and checks the validity of each variant by counting the number of appearance of the variant in the retrieved document and calculating the similarity of the context of the variant. Experiment result showed that querying with the variants produced at the first step, which is a base method of the test, performed 38% in average F measure, and querying with the refined variants at the second step, which is a proposed method, significantly improved the performance to 81% in average F measure.

• Journal of Digital Contents Society
• /
• v.18 no.7
• /
• pp.1419-1425
• /
• 2017

The ultimate goal of legal knowledge search is to obtain optimal legal information based on laws and precedent. Text mining research is actively being undertaken to meet the needs of efficient retrieval from large scale data. A typical method is to use a word embedding algorithm based on Neural Net. This paper demonstrates how to search relevant information, applying Korean law information to word embedding. First, we extracts reference laws from precedents in order and takes reference laws as input of Law2Vec. The model learns a law by predicting its surrounding context law. The algorithm then moves over each law in the corpus and repeats the training step. After the training finished, we could infer the relationship between the laws via the embedding method. The search performance was evaluated based on precision and the recall rate which are computed from how closely the results are associated to the search terms. The test result proved that what this paper proposes is much more useful compared to existing systems utilizing only keyword search when it comes to extracting related laws.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.25 no.3
• /
• pp.101-118
• /
• 2014

Given the context of rapid developing ubiquitous computing environment, it is imperative for users to search and use images based on affective meanings. However, it has been difficult to index affective meanings of image since emotions of image are substantially subjective and highly abstract. In addition, utilizing low level features of image for indexing affective meanings of image has been limited for high level concepts of image. To facilitate the access points of affective meanings of image, this study aims to utilize user-provided responses of images. For a data set, emotional words are collected and cleaned from twenty participants with a set of fifteen images, three images for each of basic emotions, love, sad, fear, anger, and happy. A total of 399 unique emotion words are revealed and 1,093 times appeared in this data set. Through co-word analysis and network analysis of emotional words from users' responses, this study demonstrates expanded word sets for five basic emotions. The expanded word sets are characterized with adjective expression and action/behavior expression.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.6
• /
• pp.106-109
• /
• 1997

For the practical isolated word recognition system, the ability to reject the out-of -vocabulary(OOV) is required. In this paper, we present a rejection method which uses the clustered phoneme modeling combined with postprocessing by likelihood ratio scoring. Our baseline speech recognition system was based on the whole-word continuous HMM. And 6 clustered phoneme models were generated using statistical method from the 45 context independent phoneme models, which were trained using the phonetically balanced speech database. The test of the rejection performance for speaker independent isolated words recogntion task on the 22 section names shows that our method is superior to the conventional postprocessing method, performing the rejection according to the likelihood difference between the first and second candidates. Furthermore, this clustered phoneme models do not require retraining for the other isolated word recognition system with different vocabulary sets.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.9
• /
• pp.1702-1709
• /
• 1994

In the postprocessing of Hangul recognition system, the storage structure of contextual information is an important matter for the recognition rate and speed of the entire system. Trie in general is used to represent the context as word dictionary, but the memory space efficiency of the structure is low. Therefore we propose a new structure for word dictionary that has better space efficiency and the equivalent merits of trie. Because Hangul is a compound language, the language can be represented by phonemes or by characters. In the representation by phonemes(P-mode) the retrieval is fast, but the space efficiency is low. In the representation by characters(C-mode) the space efficiency is high, but the retrieval is slow. In this paper the two representation methods are combined to form a hybrid representation(H-mode). At first an optimal level for the combination is selected by two characteristic curves of node utilization and dispersion. Then the input words are represented with trie structure by P-mode from the first to the optimal level, and the rest are represented with sequentially linked list structure by C-mode. The experimental results for the six kinds of word set show that the proposed structure is more efficient. This result is based on the fact that the retrieval for H-mode is as fast as P-mode and the space efficiency is as good as C-mode.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.6
• /
• pp.11-23
• /
• 2002

Computational models up to now for Korean morphology have been linear in that it deal with only segmentation of morphemes rather than formation of the internal structure of a word. When integrating a linear computational model with syntax analysis, it requires an additional interface component between this model and the syntax to bind morphemes into sentence constituents. Furthermore the linear model is not semantically intuitive. In this paper, based on word-syntactical viewpoint, we propose an integrated computational model that deals with morpheme segmentation, formation of syntactic element (sentence constituent), and even internal structure of word. Formalism of two-level morphology is employed to cope with morpheme segmentation and alternation problems, and functional diacritics are proposed to incorporate categorial context into the two-level formalism. A modified GLR-based algorithm is also proposed to check syntactical constraint of morphemes.