• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1639-1658
• /
• 2019

Named entity recognition (NER) is an important technique for improving the performance of data mining and big data analytics. In previous studies, NER systems have been employed to identify named-entities using statistical methods based on prior information or linguistic features; however, such methods are limited in that they are unable to recognize unregistered or unlearned objects. In this paper, a method is proposed to extract objects, such as technologies, theories, or person names, by analyzing the collocation relationship between certain words that simultaneously appear around specific words in the abstracts of academic journals. The method is executed as follows. First, the data is preprocessed using data cleaning and sentence detection to separate the text into single sentences. Then, part-of-speech (POS) tagging is applied to the individual sentences. After this, the appearance and collocation information of the other POS tags is analyzed, excluding the entity candidates, such as nouns. Finally, an entity recognition model is created based on analyzing and classifying the information in the sentences.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.91-94
• /
• 2021

As deep learning technologies becoming developed, realistic fake videos synthesized by deep learning models called "Deepfake" videos became even more difficult to distinguish from original videos. As fake news or Deepfake blackmailing are causing confusion and serious problems, this paper suggests a novel model detecting Deepfake videos. We chose Residual Convolutional Neural Network (Resnet50) as an extraction model and Long Short-Term Memory (LSTM) which is a form of Recurrent Neural Network (RNN) as a classification model. We adopted cosine similarity with hinge loss to train our extraction model in embedding the features of Deepfake and original video. The result in this paper demonstrates that temporal features in the videos are essential for detecting Deepfake videos.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.11
• /
• pp.1031-1040
• /
• 2014

In this paper, a new condition diagnosis algorithm for the lens injection molding process using various features extracted from cavity pressure, nozzle pressure and screw position signals is developed with the aid of probability neural network (PNN) method. A new feature extraction method is developed for identifying five (5), seven (7) and two (2) critical features from cavity pressure, nozzle pressure and screw position signals, respectively. The node energies extracted from cavity and nozzle pressure signals are also considered based on wavelet packet decomposition (WPD). The PNN method is introduced to build the condition diagnosis model by considering the extracted features and node energies. A series of the lens injection molding experiments are conducted to validate the model, and it is demonstrated that the proposed condition diagnosis model is useful with high diagnosis accuracy.

• Journal of the Korean Society of Industry Convergence
• /
• v.6 no.1
• /
• pp.23-29
• /
• 2003

This paper proposes an efficient feature extraction of the images by using independent component analysis(ICA) based on neural networks of the hybrid learning algorithm. The proposed learning algorithm is the fixed point(FP) algorithm based on Newton method and moment. The Newton method, which uses to the tangent line for estimating the root of function, is applied for fast updating the inverse mixing matrix. The moment is also applied for getting the better speed-up by restraining an oscillation due to compute the tangent line. The proposed algorithm has been applied to the 10,000 image patches of $12{\times}12$-pixel that are extracted from 13 natural images. The 144 features of $12{\times}12$-pixel and the 160 features of $16{\times}16$-pixel have been extracted from all patches, respectively. The simulation results show that the extracted features have a localized characteristics being included in the images in space, as well as in frequency and orientation. And the proposed algorithm has better performances of the learning speed than those using the conventional FP algorithm based on Newton method.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.9
• /
• pp.758-765
• /
• 2001

This paper proposes a method to automatically extract stable visual landmarks from sensory data. Given a 2D occupancy map, a mobile robot first extracts vertical line features which are distinct and on vertical planar surfaces, because they are expected to be observed reliably from various viewpoints. Since the feature information such as position and length includes uncertainty due to errors of vision and motion, the robot then reduces the uncertainty by matching the planar surface containing the features to the map. As a result, the robot obtains modeled stable visual landmarks from extracted features. This extraction process is performed on-line to adapt to an actual changes of lighting and scene depending on the robot’s view. Experimental results in various real scenes show the validity of the proposed method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.3
• /
• pp.398-403
• /
• 2006

In this paper, we describe a fuzzy threshold selection technique for feature extraction in digital color images. this is achieved by the formulation a fuzzy inference system that evaluates threshold for feature configurations. The system uses two fuzzy measures. They capture desirable characteristics of features such as dependency of local intensity and continuity in an image. We give a graphical description of a nonlinear sketch feature extraction filter and design the fuzzy inference system in terms of the characteristics of the feature. Through the design, we provide selection method on the choice of a threshold to achieve certain characteristics of the extracted features. Experimental results show the usefulness of our fuzzy threshold inference approach which is able to extract features without human intervention.

• Journal of the Institute of Convergence Signal Processing
• /
• v.1 no.1
• /
• pp.42-48
• /
• 2000

In this paper, we propose the face image recognition algorithm using skin color information, face region features such as eye, nose, and mouse, etc., and geometrical features of chin line. In the proposed algorithm, we used the intensity as well as skin color information in the HSI color coordinate which is similar to human eye system. The experimental results of proposed method shows improved extraction quality of face and provides adaptive extraction methods for the races. And also, we used chin line information as well as geometrical features of face such as eye, nose, mouse information for the improvement of face recognition quality, Experimental results shows the more improved recognition as well as extraction quality than conventional methods.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.5 s.43
• /
• pp.53-58
• /
• 2006

In this paper, I propose an image based feature extraction method for an automatic golf swing analysis. While most swing analysis systems require an expert like teaching professional, the proposed method enables an automatic swing analysis without a professional. The extracted features for swing analysis include not only key frames such as addressing, backward swing, top, forward swing, impact, and follow-through swing but also important positions of golfer's body parts such as hands, shoulders, club head, feet, knee. To see the effectiveness of the proposed method. I tested it for several swing pictures. Experimental results show that the proposed method is effective for extracting important swing features. Further research is under going to develop an automatic swing analysis system using the proposed features.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.2
• /
• pp.348-358
• /
• 1996

If a method of character region extraction will have general purposes, it could not but make use of common features which all target images have. This paper suggests these common features should be considered as the coalitions for the region to be extracted within a framework of the cost minimization. The method suggested above could be effective by minimizing a cost function estmating the extent that character regions satify quantitatively the features, through Simulated Annealing Method. This method has an uniqueness in that it defines the cost function. Experimental result verify the usefulness of this cost minimization approach to characer region extraction.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.226-228
• /
• 2005

Predicting the cellular location of an unknown protein gives valuable information for inferring the possible function of the protein. For more accurate Prediction system, we need a good feature extraction method that transforms the raw sequence data into the numerical feature vector, minimizing information loss. In this paper we propose new methods of extracting underlying features only from the sequence data by computing pairwise sequence alignment scores. In addition, we use composition based features to improve prediction accuracy. To construct an SVM ensemble from separately trained SVM classifiers, we propose specificity based weighted majority voting . The overall prediction accuracy evaluated by the 5-fold cross-validation reached $88.53\%$ for the eukaryotic animal data set. By comparing the prediction accuracy of various feature extraction methods, we could get the biological insight on the location of targeting information. Our numerical experiments confirm that our new feature extraction methods are very useful forpredicting subcellular localization of proteins.