• Journal of KIISE:Software and Applications
• /
• v.35 no.5
• /
• pp.335-346
• /
• 2008

Causal relations in ontology should be defined based on the inference types necessary to solve problems specific to application as well as domain. In this paper, we present a model to define and extract causal relations for application ontology for Question-Answering (QA) on fault-diagnosis of electronic devices. Causal categories are defined by analyzing generic patterns of QA application; the relations between concepts in the corpus belonging to the causal categories are defined as causal relations. Instances of casual relations are extracted using lexical patterns in the concept definitions of domain, and extended incrementally with information from thesaurus. On the evaluation by domain specialists, our model shows precision of 92.3% in classification of relations and precision of 80.7% in identifying causal relations at the extraction phase.

• Journal of KIISE:Software and Applications
• /
• v.32 no.5
• /
• pp.378-384
• /
• 2005

In this paper, we describe a novel method of spam-filtering to improve the performance of conventional spam-filtering systems. Conventional systems filter emails by investigating words distribution in email headers or bodies. Nowadays, spammers begin making email accounts in web-based email service sites and sending emails as if they are not spams. Investigating the email accounts of those spams, we notice that there is a large difference between the automatically generated accounts and ordinaries. Based on that difference, incoming emails are classified into spam/non-spam classes. To classify emails from only account strings, we used decision trees, which have been generally used for conventional pattern classification problems. We collected about 2.15 million account strings from email service sites, and our account checker resulted in the accuracy of $96.3\%$. The previous filter system with the checker yielded the improved filtering performance.

• Journal of KIISE:Software and Applications
• /
• v.32 no.5
• /
• pp.396-405
• /
• 2005

Since sentences are the basic propositional units of text, their themes would be helpful for various tasks that require knowledge about the semantic content of text. Despite the importance of determining the theme of a sentence, however, few studies have investigated the problem of automatically assigning the theme to a sentence. Therefore, we propose a sentence theme allocation scheme based on the head-driven patterns of sentences in encyclopedia. In a serious of experiments using Dusan Dong-A encyclopedia, the proposed method outperformed the baseline of the theme allocation performance. The head-driven pattern 4, which is reconfigured based on the predicate, showed superior performance in the theme allocation with the average F-score of $98.96\%$ for the training data, and $88.57\%$ for the test data.

• Journal of KIISE
• /
• v.42 no.4
• /
• pp.487-495
• /
• 2015

Since most of information is computerized nowadays, it is extremely important to promote the security of the computerized information. However, the software itself can threaten the safety of information through many abusive methods enabled by coding mistakes. Even though the Secure Coding Guide has been proposed to promote the safety of information by fundamentally blocking the hacking methods, it is still hard to apply the techniques on other programming languages because the proposed coding guide is mainly written for C and Java programmers. In this paper, we reclassified the coding rules of the Secure Coding Guide to extend its applicability to programming languages in general. The specific coding guide adopted in this paper is the C Secure Coding Guide, announced by the Ministry of Government Administration and Home Affairs of Korea. According to the classification, we applied the rules of programming in Sprout, which is a newly proposed Korean programming language. The number of vulnerability rules that should be checked was decreased in Sprout by 52% compared to C.

• Journal of KIISE:Software and Applications
• /
• v.34 no.10
• /
• pp.929-936
• /
• 2007

The novel vision-based scheme for real-time extracting traffic parameters is proposed in this paper. Detecting and tracking of vehicle is processed at local region installed by operator. Local region is divided to segmented regions by edge and frame difference, and the segmented regions are classified into vehicle, road, shadow and headlight by statistical and geometrical features. Vehicle is detected by the result of the classification. Traffic parameters such as velocity, length, occupancy and distance are estimated by tracking using template matching at local region. Because background image are not used, it is possible to utilize under various conditions such as weather, time slots and locations. It is performed well with 90.16% detection rate in various databases. If direction, angle and iris are fitted to operating conditions, we are looking forward to using as the core of traffic monitoring systems.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.1
• /
• pp.13-20
• /
• 2016

In this paper, we propose an object recognition system that can effectively find out its category, its instance name, and several attributes from the color and depth images of an object with hierarchical feature learning. In the preprocessing stage, our system transforms the depth images of the object into the surface normal vectors, which can represent the shape information of the object more precisely. In the feature learning stage, it extracts a set of patch features and image features from a pair of the color image and the surface normal vector through two-layered learning. And then the system trains a set of independent classification models with a set of labeled feature vectors and the SVM learning algorithm. Through experiments with UW RGB-D Object Dataset, we verify the performance of the proposed object recognition system.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.7
• /
• pp.285-292
• /
• 2014

A named entity recognition method is used to improve the performance of information retrieval systems, question answering systems, machine translation systems and so on. The targets of the named entity recognition are usually PLOs (persons, locations and organizations). They are usually proper nouns or unregistered words, and traditional named entity recognizers use these characteristics to find out named entity candidates. The titles of books, movies and TV programs have different characteristics than PLO entities. They are sometimes multiple phrases, one sentence, or special characters. This makes it difficult to find the named entity candidates. In this paper we propose a method to quickly extract title named entities from news articles and automatically build a named entity dictionary for the titles. For the candidates identification, the word phrases enclosed with special symbols in a sentence are firstly extracted, and then verified by the SVM with using feature words and their distances. For the classification of the extracted title candidates, SVM is used with the mutual information of word contexts.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.3
• /
• pp.155-160
• /
• 2017

Recently, Deep learning has been used successfully to solve many recognition problems. It has many advantages over existing machine learning methods that extract feature points through hand-crafting. Deep neural networks for human activity recognition split video data into frame images, and then classify activities by analysing the connectivity of frame images according to the time. But it is difficult to apply to actual problems which has small-scale activity classes. Because this situations has a problem of overfitting and insufficient training data. In this paper, we defined 5 type of small-scale human activities, and classified them. We construct video database using 700 video clips, and obtained a classifying accuracy of 74.00%.

• Archives of design research
• /
• v.13 no.1
• /
• pp.139-148
• /
• 2000

Development of the digital technology made possible the Integrated media that is combined computer with information network technology. Integrated media is the data transferring method through the multimedia, and has been developed by the similar form with the human's information transferring system. For that reason, interface has been developed as the important theme, Human Computer Interaction, according to the formal development. Using interface is not activated yet for the lack of connection with various divisions and the system of the product development although interfaces formatic development is re-arranged and re-established in design. This is the try to construct new model of the interface by re-arrange of the factors for the information transferring form development on the ground of the mp3 service vending machine development. Interface is divided into logical I formatic and hardware I software for he systematic classification and also arranged so that anybody can understand. Consequently, presenting of the new integrated interface paradigm can make possible to suggest the fundamental framework that is the new direction of the design re-arranged by the integrated media.

• Journal of the Korean Society of Tobacco Science
• /
• v.20 no.2
• /
• pp.191-197
• /
• 1998

Discriminant analysis using near infrared spectra derived from Korean Flue-cured(KF) and American Flue-cured(AF), and also Korean Burley(KB) and American Burley(AB) tobacco was done to classify flue-cured and burley tobacco as either grown in Korea or grown in the USA. Samples were scanned in the wavelength of 400 ～ 2500 nm by near infrared analyzer(NIRSystem Co., model 6500). The discrimination equations for flue-cured and burley tobacco were developed using partial least square 2 method in Infrasoft International NIRS 3 software package. KF samples used for the development of the discrimination equations were higher contents of total sugar, crude ash and chlorine, and higher value of leaf density and brightness, but lower contents of nicotine, total nitrogen and ether extracts, and higher value of redness than those of AF samples. KB samples were higher contents of nicotine, crude ash and chlorine, but lower contents of ether extracts and value of brightness than those of AB samples. On 3 dimensional graph drawn with 3 principal component scores calculated with 3 principal component from KF and KB sample spectra, KF sample spectra were significantly different from AF, and also KB sample spectra were significantly different from AB. The discrimination equations of flue-cured and burley were developed with 3 principal component, respectively. The discrimination equations for flue-cured and burley had a standard error of 0.03 and 0.04, and a R2 of 0.88 and 0.84, respectively. The tobacco samples used for the development of discrimination equation were perfectly classified as KF and AF by flue-cured discrimination equation, and also perfectly classified KB and AB by burley discrimination equation, respectively. The correct classification rates of KF and AF samples not used for the development of discrimination equations were 9S % (828 out of 869 samples) and 98 % (98 out of 100 samples) by flue-cured discrimination equations, and KB and AB samples were 94%(345 out of 368 samples) and 100%(42 out of 42 samples) by burley discrimination equations, respectively.