• International Journal of Internet, Broadcasting and Communication
• /
• v.5 no.2
• /
• pp.5-10
• /
• 2013

We proposed a novel similarity measure using weighted difference entropy (WDE) to improve the performance of the CF system. The proposed similarity metric evaluates the entropy with a preference score difference between the common rated items of two users, and normalizes it based on the Gaussian, tanh and sigmoid function. We showed significant improvement of experimental results and environments. These experiments involved changing the number of nearest neighborhoods, and we presented experimental results for two data sets with different characteristics, and results for the quality of recommendation.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.11
• /
• pp.121-126
• /
• 2016

In recommender systems based on collaborative filtering, measuring similarity is very critical for determining the range of recommenders. Data sparsity problem is fundamental in collaborative filtering systems, which is partly solved by Jaccard coefficient combined with traditional similarity measures. This study proposes a new coefficient for improving performance of Jaccard coefficient by compensating for its drawbacks. We conducted experiments using datasets of various characteristics for performance analysis. As a result of comparison between the proposed and the similarity metric of Pearson correlation widely used up to date, it is found that the two metrics yielded competitive performance on a dense dataset while the proposed showed much better performance on a sparser dataset. Also, the result of comparing the proposed with Jaccard coefficient showed that the proposed yielded far better performance as the dataset is denser. Overall, the proposed coefficient demonstrated the best prediction and recommendation performance among the experimented metrics.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.181-188
• /
• 2006

In this paper, we propose a search algorithm for similar example-sentences in the computer-aided translation. The search for similar examples, which is a main part in the computer-aided translation, is to retrieve the most similar examples in the aspect of structural and semantical analogy for a given query from examples. The proposed algorithm is based on the Needleman-Wunsch algorithm, which is used to measure similarity between protein or nucleotide sequences in bioinformatics. If the original Needleman-Wunsch algorithm is applied to the search for similar sentences, it is likely to fail to find them since similarity is sensitive to word's inflectional components. Therefore, we use the lemma in addition to (typographical) surface information. In addition, we use the part-of-speech to capture the structural analogy. In other word, this paper proposes the similarity metric combining the surface, lemma, and part-of-speech information of a word. Finally, we present a search algorithm with the proposed metric and present pairs contributed to similarity between a query and a found example. Our algorithm shows good performance in the area of electricity and communication.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2527-2545
• /
• 2016

Measuring the similarity of given samples is a key problem of recognition, clustering, retrieval and related applications. A number of works, e.g. kernel method and metric learning, have been contributed to this problem. The challenge of similarity learning is to find a similarity robust to intra-class variance and simultaneously selective to inter-class characteristic. We observed that, the similarity measure can be improved if the data distribution and hidden semantic information are exploited in a more sophisticated way. In this paper, we propose a similarity learning approach for retrieval and recognition. The approach, termed as LDA-FEK, derives free energy kernel (FEK) from Latent Dirichlet Allocation (LDA). First, it trains LDA and constructs kernel using the parameters and variables of the trained model. Then, the unknown kernel parameters are learned by a discriminative learning approach. The main contributions of the proposed method are twofold: (1) the method is computationally efficient and scalable since the parameters in kernel are determined in a staged way; (2) the method exploits data distribution and semantic level hidden information by means of LDA. To evaluate the performance of LDA-FEK, we apply it for image retrieval over two data sets and for text categorization on four popular data sets. The results show the competitive performance of our method.

• Journal of KIISE:Databases
• /
• v.36 no.5
• /
• pp.366-373
• /
• 2009

In this paper, we propose a similarity search algorithm for image databases. One of the central problems regarding content-based image retrieval (CBIR) is the semantic gap between the low-level features computed automatically from images and the human interpretation of image content. Many search algorithms used in CBIR have used the Minkowski metric (or $L_p$-norm) to measure similarity between image pairs. However those functions cannot adequately capture the aspects of the characteristics of the human visual system as well as the nonlinear relationships in contextual information. Our new search algorithm tackles this problem by employing new similarity measures and ranking strategies that reflect the nonlinearity of human perception and contextual information. Our search algorithm yields superior experimental results on a real handwritten digit image database and demonstrates its effectiveness.

• Smart Media Journal
• /
• v.8 no.4
• /
• pp.80-84
• /
• 2019

As a result of AMI rapidly expanding and distributing its products, variety of services that utilize data on the use of electricity are increasing. In order to make these services more effective, missing metric data needs to be corrected, compensating for which Euclidean similarity is used to find customers with similar usage patterns. Throughout such a process, we propose a method for correcting missing data and provide comparison with the preceding methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.33-35
• /
• 2018

기존의 사람간의 유사성 측정 시스템은 적외선 빔이나 열 감지 영상 장치를 통해 측정하였다. 하지만 이와 같은 방법으로 측정하면 2명 이상의 객체를 분류해내는 기술은 제공하지 않는다. 이에 본 논문은 고정된 카메라를 이용하여 각 사람의 피부색과 옷차림 등의 RGB 정보를 이용한 사람 유사성 측정 기법을 제안한다. RGB카메라 영상을 통하여 객체의 RGB 히스토그램을 얻은 후 각 객체에 대해 Bhattacharyya metric, Cosine similarity, Jensen difference, Euclidean distance로 histogram similarity를 계산하여 객체 추적 및 유사성 측정을 통해 객체를 분류한다. 제안된 시스템은 C/C++를 기반으로 구현하여, 유사성 측정 성능을 평가하였다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.6
• /
• pp.735-744
• /
• 2019

One of the key factors for recognition performance in the automatic target recognition for synthetic aperture radar imagery(SAR-ATR) system is reliability of the SAR target database. To achieve optimal performance, the database should be constructed using the images obtained under the same operating condition as the SAR sensor. However, it is impractical to have the extensive set of real-world SAR images, and thus those from the electro magnetic prediction tool with 3-D CAD models are suggested as an alternative where their reliability can be always questionable. In this paper, a method for similarity assessment between target SAR images is presented inspired by the fact that a target SAR image is mainly characterized by the features of scattering centers. The method is demonstrated using a variety of examples and quantitatively measures the similarity related to reliability. Its assessment performance is further compared with that of the existing metric, structural similarity(SSIM).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3798-3814
• /
• 2022

Social recommendation algorithm can alleviate data sparsity and cold start problems in recommendation system by integrated social information. Among them, matrix-based decomposition algorithms are the most widely used and studied. Such algorithms use dot product operations to calculate the similarity between users and items, which ignores user's potential preferences, reduces algorithms' recommendation accuracy. This deficiency can be avoided by a metric learning-based social recommendation algorithm, which learns the distance between user embedding vectors and item embedding vectors instead of vector dot-product operations. However, previous works provide no theoretical explanation for its plausibility. Moreover, most works focus on the indirect impact of social friends on user's preferences, ignoring the direct impact on user's rating preferences, which is the influence of user rating preferences. To solve these problems, this study proposes a user bias drift social recommendation algorithm based on metric learning (BDML). The main work of this paper is as follows: (1) the process of introducing metric learning in the social recommendation scenario is introduced in the form of equations, and explained the reason why metric learning can replace the click operation; (2) a new user bias is constructed to simultaneously model the impact of social relationships on user's ratings preferences and user's preferences; Experimental results on two datasets show that the BDML algorithm proposed in this study has better recommendation accuracy compared with other comparison algorithms, and will be able to guarantee the recommendation effect in a more sparse dataset.

• Smart Media Journal
• /
• v.9 no.3
• /
• pp.9-17
• /
• 2020

Hippocampus is a complex brain structure embedded deep into the temporal lobe. Studies have shown that this structure gets affected by neurological and psychiatric disorders and it is a significant landmark for diagnosing neurodegenerative diseases. Hippocampus features play very significant roles in region-of-interest based analysis for disease diagnosis and prognosis. In this study, we have attempted to learn the embeddings of this important biomarker. As conventional metric learning methods for feature embedding is known to lacking in capturing semantic similarity among the data under study, we have trained deep Siamese convolutional neural network for learning metric of the hippocampus. We have exploited Gwangju Alzheimer's and Related Dementia cohort data set in our study. The input to the network was pairs of three-view patches (TVPs) of size 32 × 32 × 3. The positive samples were taken from the vicinity of a specified landmark for the hippocampus and negative samples were taken from random locations of the brain excluding hippocampi regions. We have achieved 98.72% accuracy in verifying hippocampus TVPs.