• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.301-304
• /
• 2005

For detecting an intrusion based on the anomaly of a user's activities, previous works are concentrated on statistical techniques or frequent episode mining in order to analyze an audit data. But, since they mainly analyze the average behaviour of user's activities, some anomalies can be detected inaccurately. Therefore, we propose an anomaly detection method that utilizes an associative classification for modelling intrusion detection. Finally, we proof that a prediction model built from associative classification method yields better accuracy than a prediction model built from a traditional methods by experimental results.

• Journal of KIISE:Databases
• /
• v.32 no.6
• /
• pp.567-584
• /
• 2005

Temporal data mining, the incorporation of temporal semantics to existing data mining techniques, refers to a set of techniques for discovering implicit and useful temporal knowledge from temporal data. Association rules and classification are applied to various applications which are the typical data mining problems. However, these approaches do not consider temporal attribute and have been pursued for discovering knowledge from static data although a large proportion of data contains temporal dimension. Also, data mining researches from temporal data treat problems for discovering knowledge from data stamped with time point and adding time constraint. Therefore, these do not consider temporal semantics and temporal relationships containing data. This paper suggests that temporal associative classification technique based on temporal class association rules. This temporal classification applies rules discovered by temporal class association rules which extends existing associative classification by containing temporal dimension for generating temporal classification rules. Therefore, this technique can discover more useful knowledge in compared with typical classification techniques.

• Information and Communications Magazine
• /
• v.7 no.3
• /
• pp.27-34
• /
• 1990

The $(L,\psi)$ feature description on the binary boundary air craft image is introduced of classifying 3-D object (aircraft) identification. Three types for associative matrix memories are employed and tested for their classification performance. The fast association involved in these memories can be implemented using a parallel optical matrix-vector operation. Two associative memories are based on pseudoinverse solutions and the third one is interoduced as a paralell version of a nearest-neighbor classifier. Detailed simulation results for each associative processor are provided.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.4
• /
• pp.215-223
• /
• 2018

Recently, as interest of wearable devices has increased, commercially available smart wristbands and applications have been used as a tool for personal healthy management. However most previous studies have focused on evaluating the accuracy and reliability of the technical problems of wearable devices, especially step counts, walking distance, and energy consumption measured from the smart wristbands. In this study, we propose a physical activity evaluation model using classification rules, induced from the associative classification mining approach. These rules associated with five physical activities were generated by considering activities and walking times in target heart rate zones such as 'Out-of Zone', 'Fat Burn Zone', 'Cardio Zone', and 'Peak Zone'. In the experiment, we evaluated the prediction power of classification rules and verified its effectiveness by comparing classification accuracies between the proposed model and support vector machine.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.115.3-115
• /
• 2001

In this paper, discrete-time cellular neural networks are designed in order to function as associative memories by using Hebbian learning rule and non-cloning template. The proposed method has a very simple structure to design and to learn. Weights are updated by the connection between the neuron and its neighborhood. In the simulation, the proposed method is applied to the classification of a traffic sign pattern.

• Proceedings of the Optical Society of Korea Conference
• /
• 1989.02a
• /
• pp.95-98
• /
• 1989

A new associative memory model using the SDF filter, one of the multiple filter for pattern recognition, is suggested in this paper. The SDF filter characteristics such as pattern classification lets the memorized patterns have orthogonal characteristics one another, so that enhances the associative memory's retrieval ability to the original pattern. The computer simulation shows that this new model is very useful in case that the imput patterns are seriously distorted and the cross-correlation between the memorized patterns is very high.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1395-1397
• /
• 1996

Many of the existing neural associative memories are trained and recalled in separate modes and are not suitable for temporal pattern storage and classification in that user must specify the time and length of input patterns. In this paper, a new on-line temporal associative memory model is presented. This memory is structured in layers of neurons and each neuron has limited number of weights so that calculation complexity can be considerably reduced and processing of patterns can be achieved in real time.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.2
• /
• pp.154-162
• /
• 2003

In this paper, Associative Memory with Cellular Neural Networks classifies and recognizes image patterns as an operator applied to image process. CNN processes nonlinear data in real-time like neural networks, and made by cell which communicates with each other directly through its neighbor cells as the Cellular Automata does. It is applied to the optimization problem, associative memory, pattern recognition, and computer vision. Image processing with CNN is appropriate to 2-D images, because each cell which corresponds to each pixel in the image is simultaneously processed in parallel. This paper shows the method for designing the structure of associative memory based on CNN and getting output image by choosing the most appropriate weight pattern among the whole learned weight pattern memories. Each template represents weight values between cells and updates them by learning. Hebbian rule is used for learning template weights and LMS algorithm is used for classification.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.154-156
• /
• 2003

인터넷의 급속한 성장과 더불어 많은 정보와 데이터들을 인터넷을 통하여 얻을 수 있게 되었으며 많은 단체들이 문서들을 웹을 통하여 이용 가능하게 만들고 있다. 이에 따라 다양한 정보와 데이터를 효과적으로 분류하고 검색하는 문서 분류 (Document Classification)에 대한 알고리즘이 다양한 분야에서 널리 연구되어 왔으며 본 논문에서 초점을 두고 있는 전자 도서관 (Digital Library) 분야에서도 활발히 연구되어지고 있다. 하지만 기존의 전자 도서관의 문서 분류 알고리즘들은 문서들의 각 단락의 비중을 고려하지 않은 채 단어들의 발생 빈도에 초점을 두어 많은 잡음 단어 (Noise Term)를 포함하고 그로 인하여 분류 성능이 떨어졌다. 본 논문에서는 문서 단락의 중요도에 따라 다른 .가중치를 부여하여 단어 지지도 (Term Support)가 높은 단어들을 추출하고 그 단어들로 연관 규칙 (Association Rules)을 이용하여 분류 규칙을 생성하는 방법을 제안한다. 제안된 방법의 성능평가를 위해 문서 분류에 널리 쓰이는 나이브 베이지안 분류자 (Na$\square$ve Bayesian Classifier) 및 기존의 단순 연관 규칙 분류자 (Associative Classifier)와 비교 평가하였다. 그 결과, 각 가중치가 부여된 연관 규칙 분류 방법이 나이브 베이지안 분류 방법과 단순 연관 규칙 분류 방법보다 높은 성능을 보였다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2002.05a
• /
• pp.632-638
• /
• 2002

In novelty detection, one attempts to discriminate abnormal patterns from normal ones. Novelty detection is quite difficult since, unlike usual two class classification problems, only normal patterns are available for training. Auto-Associative Multi-Layer Perceptron (AAMLP) has been shown to provide a good performance based upon the property that novel patterns usually have larger auto-associative errors. In this paper, we give a mathematical analysis of 2-layer AAMLP's output characteristics and empirical results of 2-layer and 4-layer AAMLPs. Various activation functions such as linear, saturated linear and sigmoid are compared. The 2-layer AAMLPs cannot identify non-linear boundaries while the 4-layer ones can. When the data distribution is multi-modal, then an ensemble of AAMLPs, each of which is trained with pre-clustered data is required. This paper contributes to understanding of AAMLP networks and leads to practical recommendations regarding its use.