Search

Grammatical Error Correction (GEC) is the task of detecting and correcting various grammatical errors in texts. Many previous approaches to the GEC have used various mechanisms including rules, statistics, and their combinations. Recently, the performance of the GEC in English has been drastically enhanced due to the vigorous applications of deep neural networks and pretrained language models. Following the promising results of the English GEC tasks, we apply the Transformer with Copying Mechanism into the Korean GEC task by introducing novel and effective noising methods for constructing Korean GEC datasets. Our comparative experiments showed that the proposed system outperforms two commercial grammar check and other NMT-based models.

The automatic extraction of protein–protein interactions (PPIs) reported in scientific publications are of great significance for biomedical researchers in that they could efficiently grasp the recent research results about biochemical events and molecular processes for conducting their original studies. This article introduces a deep convolutional neural network (DCNN) equipped with various feature embeddings to battle the limitations of the existing machine learning-based PPI extraction methods. The proposed model learns and optimises word embeddings based on the publicly available word vectors and also exploits position embeddings to identify the locations of the target protein names in sentences. Furthermore, it can employ various linguistic feature embeddings to improve the PPI extraction. The intensive experiments using AIMed data set known as the most difficult collection not only show the superiority of the suggested model but also indicate important implications in optimising the network parameters and hyperparameters.

A new two-line based inversion driving method is introduced for low power display-driver ICs. By inserting a timing offset between the chopper stabilization and the alternation of LCD polarity, we can reduce power consumption without noticeable degradation in the display quality. By applying the proposed scheme to 12″ LCD applications, we achieved 7.5% and 27% power saving in the display-driver IC with white and black patterns, respectively.

In maritime logistics, the technology for transmitting status information has been greatly developed, but it has not been available in general logistics environment or it is impossible to support two-way communication as it simply receives and transfers a container's information. In particular, to support two-way communication in all sections of the global maritime logistics, the address value, which can identify the tag, should be managed. In this context, to improve the 433 MHz-based RFID, the project called "DASH 7" has been conducted in recent years. However, it has stopped proceeding because of its slow progress, including the standardization and development of references. In this paper, we introduced an active IP-RFID system configuration for real-time communication in global maritime logistics using a two-way communication-which is characteristic of an IP-applying virtual address values in the RFID tag, and proposing its process and function. When you apply the IP-RFID system proposed in this paper, not only real-time status tracking in the maritime logistics area is possible, but it is also applicable for controlling the tag.

A slew-rate control method of LCD driver ICs is introduced to increase uniformity between adjacent driver ICs and reduce power consumption. The slew rate of every voltage follower is calibrated by a feedback algorithm during the non-displaying period. Under normal operation mode, the slew rate is dynamically controlled for improving power efficiency. Experimental results show that the power consumption is reduced by 16% with a white pattern and by 10% with a black pattern, and display defects are successfully eliminated.

Technical terms play an important role of effective queries for many users to search scientific databases. However, authors of scientific literature often employ alternative expressions to represent the meanings of specific terms, in other words, Terminological Paraphrases (TPs) in the literature for certain reasons, which leads to producing relevant documents that are not captured by conventional terms above. In this paper, we propose an effective way to retrieve "de facto relevant documents" which only contain those TPs and cannot be searched by conventional models in an environment with only controlled vocabularies by adapting Predicate Argument Tuple (PAT). The experiment confirms that PAT-based document retrieval is an effective and promising method to discover those kinds of documents and to improve the recall of terminology-based scientific information access models.

An essential matter in the knowledge-based information society is how to extract useful information quickly from a large volume of literature. Since most existing data mining frameworks deal with structured input data, many limitations are faced in analyzing unstructured scientific literature and extracting new information. This study proposes a scientific-knowledge processing framework, which offers high performance by using grid computing technology for extracting important entities and their relations from the scientific literature. Since the grid computing provides a large volume of data storage and high-speed computing, the proposed framework can efficiently analyze the massive body of scientific literature and process knowledge. The workflow tool that we have developed for the proposed framework enables users to easily design and execute complicated applications that consist of complicated scientific-knowledge processes. The experimental results showed that the proposed framework reduced working time by approximately 83 % when the number of running nodes was assigned in accordance with the workload ratio of each step in scientific-knowledge processes. As a result, it is possible to effectively process a large volume of scientific literature by flexibly adjusting the number of computing nodes that constitute the grid environment as the number of documents for processing increases.

Relation extraction refers to a method of efficiently detecting and identifying predefined semantic relationships within a set of entities in text documents. Numerous relation extractionfc techniques have been developed thus far, owing to their innate importance in the domain of information extraction and text mining. The majority of the relation extraction methods proposed to date is based on a supervised learning method requiring the use of learning collections; such learning methods can be classified into feature-based, semi-supervised, and kernel-based techniques. Among these methods, a case analysis on a kernel-based relation extraction method, considered the most successful of the three approaches, is carried out in this paper. Although some previous survey papers on this topic have been published, they failed to select the most essential of the currently available kernel-based relation extraction approaches or provide an in-depth comparative analysis of them. Unlike existing case studies, the study described in this paper is based on a close analysis of the operation principles and individual characteristics of five vital representative kernel-based relation extraction methods. In addition, we present deep comparative analysis results of these methods. In addition, for further research on kernel-based relation extraction with an even higher performance and for general high-level kernel studies for linguistic processing and text mining, some additional approaches including feature-based methods based on various criteria are introduced.

Due to the development of wireless devices such as smart-phone and internet, a lot of linguistic resources actively are opened in each area of expertise. Also, various systems using semantic web technologies are developing for determining whether such information are useful or not. In order to build these systems, the processes of data collection and natural language processing are necessary. But, there is few systems to use of integrating software and data required those processes. In this paper, we propose the system, uLAMP, integrating software and data related to natural language processing. In terms of economics, the cost can be reduced by preventing duplicated implementation and data collection. On the other hand, data and software usability are increasing in terms of management aspects. In addition, for the evaluation of uLAMP usability and effectiveness, user survey was conducted. Through this evaluation, the advantages of the currentness of data and the ease of use are found.

Terminological paraphrases (TPs) are sentences or phrases that express the concepts of terminologies in a different form. Here we propose an effective way to identify and extract TPs from large-scale scientific literature databases. We propose a novel method for effectively retrieving sentences that contain a given terminological concept based on semantic units called predicate-argument tuples. This method enables effective textual similarity computations and minimized errors based on six TP ranking models. For evaluation, we constructed an evaluation collection for the TP recognition task by extracting TPs from a target literature database using the proposed method. Through the two experiments, we learned that scientific literature contain many TPs that could not have been identified so far. Also, the experimental results showed the potential and extensibility of our proposed methods to extract the TPs.

For last decade, the amount of information has been increased rapidly because of the internet and computing technology development, mobile devices and sensors, and social networks like facebook or twitter. People who want to gain important knowledge from database have been frustrated with large database. Many studies for automatic knowledge extracting meaningful knowledge from large database have been fulfilled. In that sense, automatic knowledge extracting with computing technology has been highly significant in information technology field, but still has many challenges to go further. In order to improve the effectives and efficiency of knowledge extracting system, test collection is strongly necessary. In this research, we introduce a test collection for automatic knwoledge extracting. We name the test collection KEEC/KREC(KISTI Entity Extraction Collection/KISTI Relation Extraction Collection) and present the process and guideline for building as well as the features of. The main feature is to tag by experts to guarantee the quality of collection. The experts read documents and tag entities and relation between entities with a tool for tagging. KEEC/KREC is being used for a research to evaluate system performance and will continue to contribute to next researches.

This paper proposes a model for recognizing sentential paraphrases through Predicate-Argument Tuple (PAT)-based approximate alignment between two texts. We cast the paraphrase recognition problem as a binary classification by defining and applying various alignment features which could effectively express the semantic relatedness between two sentences. Experiment confirmed the potential of our approach and error analysis revealed various paraphrase patterns not being solved by our system, which can help us devise methods for further performance improvement.

Terminology recognition system which is a preceding research for text mining, information extraction, information retrieval, semantic web, and question-answering has been intensively studied in limited range of domains, especially in bio-medical domain. We propose a domain independent terminology recognition system based on machine learning method using dictionary, syntactic features, and Web search results, since the previous works revealed limitation on applying their approaches to general domain because their resources were domain specific. We achieved F-score 80.8 and 6.5% improvement after comparing the proposed approach with the related approach, C-value, which has been widely used and is based on local domain frequencies. In the second experiment with various combinations of unithood features, the method combined with NGD(Normalized Google Distance) showed the best performance of 81.8 on F-score. We applied three machine learning methods such as Logistic regression, C4.5, and SVMs, and got the best score from the decision tree method, C4.5.

In this paper, we propose a novel kernel called a semantic parse tree kernel that extends the parse tree kernel previously studied to extract protein-protein interactions(PPIs) and shown prominent results. Among the drawbacks of the existing parse tree kernel is that it could degenerate the overall performance of PPI extraction because the kernel function may produce lower kernel values of two sentences than the actual analogy between them due to the simple comparison mechanisms handling only the superficial aspects of the constituting words. The new kernel can compute the lexical semantic similarity as well as the syntactic analogy between two parse trees of target sentences. In order to calculate the lexical semantic similarity, it incorporates context-based word sense disambiguation producing synsets in WordNet as its outputs, which, in turn, can be transformed into more general ones. In experiments, we introduced two new parameters: tree kernel decay factors, and degrees of abstracting lexical concepts which can accelerate the optimization of PPI extraction performance in addition to the conventional SVM`s regularization factor. Through these multi-strategic experiments, we confirmed the pivotal role of the newly applied parameters. Additionally, the experimental results showed that semantic parse tree kernel is superior to the conventional kernels especially in the PPI classification tasks.

Large-scaled information extraction plays an important role in advanced information retrieval as well as question answering and summarization. Information extraction can be defined as a process of converting unstructured documents into formalized, tabular information, which consists of named-entity recognition, terminology extraction, coreference resolution and relation extraction. Since all the elementary technologies have been studied independently so far, it is not trivial to integrate all the necessary processes of information extraction due to the diversity of their input/output formation approaches and operating environments. As a result, it is difficult to handle scientific documents to extract both named-entities and technical terms at once. In this study, we define scientific as a set of 10 types of named entities and technical terminologies in a biomedical domain. in order to automatically extract these entities from scientific documents at once, we develop a framework for scientific core entity extraction which embraces all the pivotal language processors, named-entity recognizer, co-reference resolver and terminology extractor. Each module of the integrated system has been evaluated with various corpus as well as KEEC 2009. The system will be utilized for various information service areas such as information retrieval, question-answering(Q&A), document indexing, dictionary construction, and so on.

This paper introduces DICE, a Domain-Independent text Classification Engine. DICE is robust, efficient, and domain-independent in terms of software and architecture. Each module of the system is clearly modularized and encapsulated for extensibility. The clear modular architecture allows for simple and continuous verification and facilitates changes in multiple cycles, even after its major development period is complete. Those who want to make use of DICE can easily implement their ideas on this test bed and optimize it for a particular domain by simply adjusting the configuration file. Unlike other publically available tool kits or development environments targeted at general purpose classification models, DICE specializes in text classification with a number of useful functions specific to it. This paper focuses on the ways to locate the optimal states of a practical text classification framework by using various adaptation methods provided by the system such as feature selection, lemmatization, and classification models.

In order to achieve relevant scholarly information from the biomedical databases, researchers generally use technical terms as queries such as proteins, genes, diseases, and other biomedical descriptors. However, the technical terms have limits as query terms because there are so many indirect and conceptual expressions denoting them in scientific literatures. Combinatorial weighting schemes are proposed as an initial approach to this problem, which utilize various indexing and weighting methods and their combinations. In the experiments based on the proposed system and previously constructed evaluation collection, this approach showed promising results in that one could continually locate new relevant expressions by combining the proposed weighting schemes. Furthermore, it could be ascertained that the most outperforming binary combinations of the weighting schemes, showing the inherent traits of the weighting schemes, could be complementary to each other and it is possible to find hidden relevant documents based on the proposed methods.

In this paper, we introduce a domain-independent classification framework based on both k-nearest neighbor and Naive Bayesian classification algorithms. The architecture of our system is simple and modularized in that each sub-module of the system could be changed or improved efficiently. Moreover, it provides various feature selection mechanisms to be applied to optimize the general-purpose classifiers for a specific domain. As for the enhanced classification performance, our system provides conditional probability boosting (CPB) mechanism which could be used in various domains. In the mood classification domain, our optimized framework using the CPB algorithm showed 1% of improvement in precision and 2% in recall compared with the baseline.

This paper suggested a way to solve the problems by using domain experts who are already in the significant level of knowledge in those fields. For the purpose of achieving our goal, a very simple and efficient approach to construct the knowledge-base which can play an important role in providing researchers with essential information in need was proposed. In addition, the Adaptive R&D Information Service Portal with a new schema structure and a construction method of representing expert`s knowledge efficiently was developed. With the simplicity and expandability of the proposed system it can be a good model for a similar system to be developed.

Background Since their introduction in 2009, the BioNLP Shared Task events have been instrumental in advancing the development of methods and resources for the automatic extraction of information from the biomedical literature. In this paper, we present the Cancer Genetics (CG) and Pathway Curation (PC) tasks, two event extraction tasks introduced in the BioNLP Shared Task 2013. The CG task focuses on cancer, emphasizing the extraction of physiological and pathological processes at various levels of biological organization, and the PC task targets reactions relevant to the development of biomolecular pathway models, defining its extraction targets on the basis of established pathway representations and ontologies. Results Six groups participated in the CG task and two groups in the PC task, together applying a wide range of extraction approaches including both established state-of-the-art systems and newly introduced extraction methods. The best-performing systems achieved F-scores of 55% on the CG task and 53% on the PC task, demonstrating a level of performance comparable to the best results achieved in similar previously proposed tasks. Conclusions The results indicate that existing event extraction technology can generalize to meet the novel challenges represented by the CG and PC task settings, suggesting that extraction methods are capable of supporting the construction of knowledge bases on the molecular mechanisms of cancer and the curation of biomolecular pathway models. The CG and PC tasks continue as open challenges for all interested parties, with data, tools and resources available from the shared task homepage.

In the present study, we develop an analysis module to be applicable to design of sliding thrust bearings. The pressure equation is solved by using the finite element method. Average lubricant temperature is obtained from using the energy balance method. The module developed has been applied to three types of thrust bearing, such as tapered-land thrust bearings of angular and diamond types, and tilting-pad thrust bearings. Effects of the dam of the tapered-lad thrust bearings have also been investigated. It has been seen that the tapered-land thrust bearings of angular type result in the highest load capacity, while the tilting pad thrust bearings result in the lowest lubricant temperature. It has also been seen that the dam in the tapered-land thrust bearings increases both the load capacity and lubricant temperature.

In this paper, we developed Korean Morphological Analyzer for an automatic indexing that is essential for Information Retrieval. Since it is important to index large-scaled document set efficiently, we concentrated on maximizing the speed of word analysis, modularization and structuralization of the system without new concepts or ideas. In this respect, our system is characterized in terms of software engineering aspect to be used in real world rather than theoretical issues. First, a dictionary of words was structured. Then modules that analyze substantive words and inflected words were introduced. Furthermore numeral analyzer was developed. And we introduced an unknown word analyzer using the patterns of morpheme. This whole system was integrated into K-2000, an information retrieval system.

One of the key issues for providing users user-customized or context-aware services is to automatically detect latent topics, users’ interests, and their changing patterns from large-scale social network information. Most of the current methods are devoted either to discovering static latent topics and users’ interests or to analyzing topic evolution only from intrafeatures of documents, namely, text content, without considering directly extrafeatures of documents such as authors. Moreover, they are applicable only to the case of single processor. To resolve these problems, we propose a dynamic users’ interest discovery model with distributed inference algorithm, named as Distributed Author-Topic over Time (D-AToT) model. The collapsed Gibbs sampling method following the main idea of MapReduce is also utilized for inferring model parameters. The proposed model can discover latent topics and users’ interests, and mine their changing patterns over time. Extensive experimental results on NIPS (Neural Information Processing Systems) dataset show that our D-AToT model is feasible and efficient.

This paper introduces InSciTe Adaptive, which is an R&D decision support system for strategic foresight, developed by KISTI. InSciTe Adaptive supports technology intelligence & competitive intelligence based on various resources such as paper, patents, and Web news. It includes eight services: Technology Navigation, Technology Trends, Core Elementary Technology, Convergence Technology, Agent Level, Agent Partner, Integrated Roadmap, and InSciTe Adaptive Report. We expect that users can acquire strategic foresight on their technology fields.

Technology trend analysis systems use data mining to process vast amounts of papers, patents and news articles to analyze and predict the life cycles of technologies, products and other kinds of entities. Some systems can also extract relations between entities such as technologies, authors and products. In order to establish precise relations between entities, entity disambiguation has to be performed. In this study, we focused on author disambiguation in the context of technology trend analysis. We used Random Forests and SVM to learn a pair wise similarity function to decide whether two articles were written by the same author or not. Besides comparing common features such as article titles and author affiliations we also studied features that were built from the analyses that were made by KISTI's InSciTe system. For training and evaluation a corpus containing 24, 750 pair wise article similarities was manually constructed using data from InSciTe. Using this corpus, Random Forests outperformed SVM and reached an accuracy value of 98.31%. Only using the newly introduced features, an accuracy of 94.79% was achieved, proving their usefulness.

From the view of design claims for online communities, it is very crucial to take interactions among members in a community into account when starting and maintaining it. This means managers of online communities need to technically support their members through online discussion and commenting services. Online discussion and commenting service, so called, blog comment hosting service, helps communities to provide their members with feedbacks of others, since such feedbacks play much important role in starting and maintaining an online community. Through online discussion and commenting services, we can post a comment on the website using our own social network service account if the website uses a social comment platform. Whenever, whatever, and wherever users post a comment, every comment is integrated and managed by the social comment platform. One of most powerful social comment platforms is Disqus. It is the social comments platform or social discussion platform used in the world popular websites such as CNN, Billboard. Thus, we analyze it in various views and give a several suggestions to make the websites more active. Main findings reported in this paper include significant implications on the design of social comment platforms.

Numerous linguistic resources are readily available in area of expertise due to the development of wireless devices such as smart-phones and the internet. To select useful information from the massive amount of the data, many systems using semantic web technologies have been developed. In order to build those systems, data collection and natural language processing are essential. However, most of those systems do not consider software integration and the data required by the processes used. In this paper, we propose a system, entitled uLAMP which integrates software and data related to natural language processing. In terms of economics, the cost is reduced by preventing duplicated implementation and data collection. On the other hand, data and software usability are increasing in terms of management requirements. In addition, for the evaluation of the uLAMP usability and effectiveness of uLAMP, a user survey was conducted. Through this evaluation, the advantages of the currentness of data and the ease of use were found.

Construction and maintenance of signaling pathway is a time-consuming and labor-intensive task. In addition, integration of various pathways is also ineffective since several markup languages are used to express pathways. To overcome these limitation, automatic pathway construction and extension with a standard format may provide a solution. The proposed approach has constructed a gold standard corpus that describes the signaling pathways, and it has been used to training and evaluating the automatic pathway construction and extension. Moreover, a standard format to express the signaling pathways has been developed and has been used to express the previous major 10 signaling pathways. An effective visualization tool has been also developed for the standardized format as well. The visualization tool can help to construct pathways and extend the current pathways using all articles in PubMed.

A new two-line based inversion driving method is introduced for low-power consumption of display driver ICs. Eliminating the correlation between the chopper stabilization and the polarity of LCD, we can reduce power consumption without noticeable degradation of display quality. Applying the proposed scheme to 17″ LCD applications, we achieved 12% power saving in DDI with white patterns and 27% with black patterns.

Many corpus-based statistical methods have been used to tackle issues of extracting disease-gene relations (DGRs) from literature. There are two limitations in the corpus-based approach: One is that available corpora for training a system are not enough and the other is that previous most research have not deal with various types of DGRs but a binary relation. In other words, analysis of presence of relation itself has been a common issue. However, the binary relation is not enough to explain DGR in practice. One solution is to construct a corpus that can analyze various types of relations between diseases and their related genes. This article describes a corpus construction process with respect to the DGRs. Eleven topics of relations were defined by biologists. Four annotators participated in the corpus annotation task and their inter-annotator agreement was calculated to show reliability for the annotation results. The gold standard data in the proposed approach can be used to enhance the performance of many research. Examples include recognition of gene and disease names and extraction of fine-grained DGRs. The corpus will be released through the GENIA project home page.

Chronic alcohol consumption contributes to numerous diseases, including cancers, cardiovascular diseases, and liver cirrhosis. Epidemiological studies have shown that excessive alcohol consumption is a risk factor for dementia. Along this line, Alzheimer's disease (AD) is the most common form of dementia and is caused by the accumulation of amyloid-β (Aβ plaques in neurons. In this study, we hypothesized that chronic ethanol consumption is associated with pathological processing of APP in AD. To investigate the relationship between chronic alcohol consumption and Aβ production, brain samples from rats fed an alcohol liquid diet for 5 weeks were analyzed. We show that the expression levels of APP, BACE1, and immature nicastrin were increased in the cerebellum, hippocampus, and striatum of the alcohol-fed group compared to the control group. Total nicastrin and PS1 levels were induced in the hippocampus of alcohol-fed rats. These data suggest that the altered expression of APP and Aβ-producing enzymes possibly contributes to the chronic alcohol consumption-mediated pathogenesis of AD.

Lots of valuable textual information is used to extract relations between named entities from literature. Composite kernel approach is proposed in this paper. The composite kernel approach calculates similarities based on the following information: (1) Phrase structure in convolution parse tree kernel that has shown encouraging results. (2) Predicate-argument structure patterns. In other words, the approach deals with syntactic structure as well as semantic structure using a reciprocal method. The proposed approach was evaluated using various types of test collections and it showed the better performance compared with those of previous approach using only information from syntactic structures. In addition, it showed the better performance than those of the state of the art approach.

We propose how to extract procedural knowledge rather than declarative knowledge utilizing machine learning method with deep language processing features in scientific documents, as well as how to model it. We show the representation of procedural knowledge in PubMed abstracts and provide experiments that are quite promising in that it shows 82%, 63%, 73%, and 70% performances of purpose/solutions (two components of procedural knowledge model) extraction, process's entity identification, entity association, and relation identification between processes respectively, even though we applied strict guidelines in evaluating the performance.

Terminology recognition system which is a fundamental research for Technology Opportunity Discovery (TOD) has been intensively studied in limited range of domains, especially in bio-medical domain. We propose a domain independent terminology recognition system based on machine learning method using dictionary, syntactic features, and Web search results, since the previous works revealed limitation on applying their approaches to general domain because their resources were domain specific. We achieved F-score 80.4 and 6.4% improvement after comparing the proposed approach with the related approach, C-value, which has been widely used and is based on local domain frequencies. In the second experiment with various combinations of unithood features, the method combined with NGD(Normalized Google Distance) showed the best performance of 81.5 on F-score. We applied two machine learning methods such as Logistic regression and SVMs, and got the best score at SVMs method.

As the design rule of wafer has been shrinking, the patterns on the mask also need to be getting smaller and even smaller for some sub-resolution assist features, which makes mask inspection process need a high resolution (HR) inspection systems. For this HR mask inspection, most mask inspector makers adopt a TDI(Time Delay & Integration) sensor to enhance acquired image quality with the acceptable scan speed, thus, to minimize the inspection cost. However, even TDI sensor may not get a sufficient gray level of pattern image for the most advanced mask patterns. Furthermore, it might generate some false defects depending on the pattern shape and scan direction (in combination with pattern direction). We manufactured two programmed defect masks (PDM); one is a ArF EPSM and another is a EUV mask. By inspecting these masks with perpendicular scan directions, respectively, we evaluated the correlation between scan direction and defect size/shape experimentally. We found that the inspection with the parallel direction to pattern direction can increase the inspectability for the patterns and the defect sensitivity since this helps to enhance signal to noise ratio from the TDI sensor. Our analysis can increase sensitivity of TDI sensor effectively without any additional hardware modification.

BitTorrent (BT) is one of the most popular Peer-to- Peer (P2P) protocols for delivering media files in the Internet today. Although BT is quite efficient for sharing and downloading files by using P2P swarming technique, the users have to download almost the whole media file before playing it. This is determined by the Rarest-Block-Download-First strategy of standard BT implementations, which is designed for fast delivery of files in the systems but not for streaming application. In this paper, we present LiveBT, a new protocol which supports video-on- demand streaming service and is totally compatible to the current BitTorrent protocol. LiveBT enables users to play hot movies shared in the BT systems smoothly just after 2-3 minutes of buffering time. We also develop the prototype of LiveBT and test the performance through the real BT download tasks of media files. By comparing our prototype with some popular BT clients claiming to support view-as- download service such as Bitcomet, we find that LiveBT spends a much shorter buffering time to play and achieves quite smooth playback performance.

With the development of internet technologies, internet has been more complexly consisted of a large amount of Web documents, science technology documents, data-base and etc. Distributed retrieval system is more required to support effective retrieval and management about a large amount of Web documents, science technology documents and etc. Distributed retrieval system has to support for user to search quickly and exactly. A distributed retrieval system has to support for DB manager to manage easily. So we have developed the distributed retrieval system called dKRISTAL which finds indexing files and manages database system in real time. We have developed new dKRISTAL system which can support searching and managing database. We measured the integrated search speed of distributed retrieval system. Also this system effectively manages documents at realtime. This paper made an experiment using a large mount of science technology information system and Web documents using dKRISTAL. This paper analyzed the result of experiment.

In this paper, we suggest a process called Listener that support so that client can manage easily information retrieval system(IRS) and apply this to KRISTAL-Information Retrieval and Management System(IRMS). Usually, in relational database management system(RDBMS) such as Oracle, there are many applications and tools that client may manage system and its data. On the other hand, in information retrieval system applications for management has been performed in server-side due to the structure and purpose of the IR systems, it is difficult to manage the system. Using the proposed Listener, administrator can achieve control of search daemon process, database structure alteration, data management on client computer. This process has been designed based on XML considering scalability and readability and supports the client API by C and Java and make developer can do easily maintenance of the application for management database.

In this paper, design for EMC of LCD driver IC is proposed. By the analysis with package parasitic parameters and EMC test pattern, local power/ground noise in LCD driver IC simulated. As a result of analysis, design for lower peak of EMI spectrum and design for lower mean level of that is proposed at the same time. In addition, decoupling cap with CMOS process can decrease EMI spectrum without any further mask. The analysis of the proposed design shows that LCD driver IC can be alternatives for TFT-LCD EMC solution. It is because LCD driver IC is nearly irrelevant to LCD module whereas source PCB should be redesigned separately according to the size and resolution of LCD panel. The experimental result shows that the proposed design methodology for LCD driver IC can successfully decrease EMI spectrum of LCD module.

Sensor data is structured and generally lacks of meaning by itself, but life-logging data (time, location, etc.) out of sensor data can be utilized to create lots of meaningful information combined with social data from social networks like Facebook and Twitter. There have been many platforms to produce meaningful information and support human behavior and context-awareness through integrating diverse mobile, social, and sensing input streams. The problem is that these platforms do not guarantee the performance in terms of the processing time and even let the accuracy of output data be addressed by new studies in each area where the platform is applied. Thus, this study proposes an improved platform which builds a knowledge base for context awareness by applying distributed and parallel computing approach considering the characteristics of sensor data that is collected and processed in real-time, and compares the proposed platform with existing platforms in terms of performance. The experiment shows the proposed platform is an advanced platform in terms of processing time. We reduce the processing time by 40% compared with existing platform. The proposed platform also guarantees the accuracy compared with existing platform.