• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.11a
• /
• pp.516-525
• /
• 2005

Case-based reasoning (CBR) has been widely used in various areas due to its convenience and strength in complex problem solving. Generally, in order to obtain successful results from CBR, effective retrieval of useful prior cases for the given problem is essential. However, designing a good matching and retrieval mechanism for CBR systems is still a controversial research issue. Most prior studies have tried to optimize the weights of the features or selection process of appropriate instances. But, these approaches have been performed independently until now. Simultaneous optimization of these components may lead to better performance than in naive models. In particular, there have been few attempts to simultaneously optimize the weight of the features and selection of the instances for CBR. Here we suggest a simultaneous optimization model of these components using a genetic algorithm (GA). We apply it to a customer classification model which utilizes demographic characteristics of customers as inputs to predict their buying behavior for a specific product. Experimental results show that simultaneously optimized CBR may improve the classification accuracy and outperform various optimized models of CBR as well as other classification models including logistic regression, multiple discriminant analysis, artificial neural networks and support vector machines.

• Journal of Information Management
• /
• v.41 no.2
• /
• pp.31-46
• /
• 2010

Now named-entity recognition(NER) as a part of information extraction has been used in the fields of information retrieval as well as question-answering systems. Unlike words, named-entities(NEs) are generated and changed steadily in documents on the Web, newspapers, and so on. The NE generation causes an unknown word problem and makes many application systems with NER difficult. In order to alleviate this problem, this paper proposes a new feature generation method for machine learning-based NER. In general features in machine learning-based NER are related with words, but entities in named-entity dictionaries are related to phrases. So the entities are not able to be directly used as features of the NER systems. This paper proposes an encoding scheme as a feature generation method which converts phrase entities into features of word units. Futhermore, due to this scheme, entities with semantic information in WordNet can be converted into features of the NER systems. Through our experiments we have shown that the performance is increased by about 6% of F1 score and the errors is reduced by about 38%.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.3
• /
• pp.314-319
• /
• 2019

Computing chromagram similarity is indispensable in constructing cover song identification system. This paper proposes a code-based chromagram similarity to reduce the computational and the storage costs for cover song identification. By learning a song-specific codebook, a chromagram sequence is converted into a code sequence, which results in the reduction of the feature storage cost. We build a lookup table over the learned codebooks to compute chromagram similarity efficiently. Experiments on two music datasets were performed to compare the proposed code-based similarity with the conventional one in terms of cover song search accuracy, feature storage, and computational cost.

• Journal of Internet Computing and Services
• /
• v.12 no.6
• /
• pp.117-128
• /
• 2011

In this paper, we present a novel approach for contents-based medication image retrieval from a medication image database using the shape classification and color information of the medication. One major problem in developing a contents-based drug image retrieval system is there are too many similar images in shape and color and it makes difficult to identify any specific medication by a single feature of the drug image. To resolve such difficulty in identifying images, we propose a hybrid approach to retrieve a medication image based on shape and color features of the medication. In the first phase of the proposed method we classify the medications by shape of the images. In the second phase, we identify them by color matching between a query image and preclassified images in the first phase. For the shape classification, the shape signature, which is unique shape descriptor of the medication, is extracted from the boundary of the medication. Once images are classified by the shape signature, Hue and Saturation(HS) color model is used to retrieve a most similarly matched medication image from the classified database images with the query image. The proposed system is designed and developed especially for specific population- seniors to browse medication images by using visual information of the medication in a feasible fashion. The experiment shows the proposed automatic image retrieval system is reliable and convenient to identify the medication images.

• The Journal of the Korea Contents Association
• /
• v.6 no.5
• /
• pp.20-28
• /
• 2006

This paper proposes visual-feature extraction for each band in wavelet domain with both spatial frequency features and multi resolution features, and the combination of visual features using fuzzy integral. In addition, it uses color feature expression method taking advantage of the frequency of the same color after color quantization for reducing quantization error, a disadvantage of the existing color histogram intersection method. Also, it is found that the final similarity can be represented in a linear combination of the respective factors(Homogram, color, energy) when each factor is independent one another. With respect to the combination patterns the fuzzy measurement is defined and the fuzzy integral is taken. Experiments are peformed on a database containing 1,000 color images. The proposed method gives better performance than the conventional method in both objective and subjective performance evaluation.

• Journal of IKEEE
• /
• v.14 no.4
• /
• pp.332-338
• /
• 2010

Usually, mage feature extraction methods are performed as a pre-processing step in many applications including image retrieval, object recognition, and image indexing. Especially, in the image texture analysis, texture feature extraction methods attempt to increase texture contrast to make it easier to extract the texture features from the image. One of the distinct textures in microscopic skin image is the wrinkle, and its features could provide various useful information for the age-related applications. In this paper, we propose a scheme to extract age-related features from the skin images and improve its accuracy in the skin age estimation.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.137-145
• /
• 2006

A user remembers a melody as not the combination of pitch and duration which is written in score but the contour which is composed of the relative pitch and duration. Because of the way of remembering a melody the previous Music Information Retrieval Systems which uses keyboard Playing or score as the main input melody are not easily acceptable in Query By Humming Systems. In this paper, we mention about the considerable checkpoints for Query By Humming System and previous researches. And we propose the feature extraction which is similar with the way of remembering a melody and similarity computation algorithms between melody in humming and melody in music. The proposed similarity computation algorithms solves the problem which can be happened when only uses the relative pitches by using relative durations.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.4
• /
• pp.1033-1043
• /
• 2000

Recently as the user request on NoD (News-on-Demand) is largely increasing, there are a lot of researches to fulfill it. However, because of short life-cycle of new video data and periodical change of video data depending on anchor, it is difficult to apply the conventional video storage techniques to NOD applications directly. For this, we design and implement low-level storage manager for efficient storage and retrieval of multimedia data in NOD Services. Our low-level storage manager not only efficiently sotres video stream dat of new video itself, but also handles its index information. It provides an inverted file method for efficient text-based retrieval and an X-tree index structure for high-dimensional feature vectors. In addition, our low-level storage manager provides some application program interfaces (APIs) for storing video objects itself and index information extracted from hierarchial new video and some APIs for retrieving video objects easily by using cursors. Finally, we implement our low-level storage manager based on SHORE (Scalable Heterogeneous Object REpository) storage system by sunig a standard C++ language under UNIX operating system.

• The Journal of Korea Robotics Society
• /
• v.8 no.2
• /
• pp.82-91
• /
• 2013

This paper proposes a building recognition algorithm using watershed image segmentation algorithm and integrated region matching (IRM). To recognize a building, a preprocessing algorithm which is using Gaussian filter to remove noise and using canny edge extraction algorithm to extract edges is applied to input building image. First, images are segmented by watershed algorithm. Next, a region adjacency graph (RAG) based on the information of segmented regions is created. And then similar and small regions are merged. Second, a color distribution feature of each region is extracted. Finally, similar building images are obtained and ranked. The building recognition algorithm was evaluated by experiment. It is verified that the result from the proposed method is superior to color histogram matching based results.

• Journal of the Korean Society for information Management
• /
• v.19 no.3
• /
• pp.5-30
• /
• 2002

In this paper, classification feature is selected with focus of musical content, note sequences pattern, and measures similarity between note sequences followed by constructing clusters by similar note sequences, which is easier for users to search by showing the similar note sequences with the search result in the CBMR system. Experimental document was $\ulcorner$A Dictionary of Musical Themes$\lrcorner$, the index of theme bar focused on classical music and obtained kern-type file. Humdrum Toolkit version 1.0 was used as note sequences treat tool. The hierarchical clustering method is by stages focused on four-type similarity matrices by whether the note sequences segmentation or not and where the starting point is. For the measurement of the result, WACS standard is used in the case of being manual classification and in the case of the note sequences starling from any point in the note sequences, there is used common feature pattern distribution in the cluster obtained from the clustering result. According to the result, clustering with segmented feature unconnected with the starting point Is higher with distinct difference compared with clustering with non-segmented feature.