• The Mathematical Education
• /
• v.63 no.2
• /
• pp.165-186
• /
• 2024

Despite the growing attention on artificial intelligence-based automated scoring technology as a support method for the introduction of descriptive items in school environments and large-scale assessments, there is a noticeable lack of foundational research in mathematics compared to other subjects. This study developed an automated scoring model for two descriptive items in first-year middle school mathematics using the Random Forest algorithm, evaluated its performance, and explored ways to enhance this performance. The accuracy of the final models for the two items was found to be between 0.95 to 1.00 and 0.73 to 0.89, respectively, which is relatively high compared to automated scoring models in other subjects. We discovered that the strategic selection of the number of evaluation categories, taking into account the amount of data, is crucial for the effective development and performance of automated scoring models. Additionally, text preprocessing by mathematics education experts proved effective in improving both the performance and interpretability of the automated scoring model. Selecting a vectorization method that matches the characteristics of the items and data was identified as one way to enhance model performance. Furthermore, we confirmed that oversampling is a useful method to supplement performance in situations where practical limitations hinder balanced data collection. To enhance educational utility, further research is needed on how to utilize feature importance derived from the Random Forest-based automated scoring model to generate useful information for teaching and learning, such as feedback. This study is significant as foundational research in the field of mathematics descriptive automatic scoring, and there is a need for various subsequent studies through close collaboration between AI experts and math education experts.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.59-83
• /
• 2018

With the increasing importance of sentiment analysis to grasp the needs of customers and the public, various types of deep learning models have been actively applied to English texts. In the sentiment analysis of English texts by deep learning, natural language sentences included in training and test datasets are usually converted into sequences of word vectors before being entered into the deep learning models. In this case, word vectors generally refer to vector representations of words obtained through splitting a sentence by space characters. There are several ways to derive word vectors, one of which is Word2Vec used for producing the 300 dimensional Google word vectors from about 100 billion words of Google News data. They have been widely used in the studies of sentiment analysis of reviews from various fields such as restaurants, movies, laptops, cameras, etc. Unlike English, morpheme plays an essential role in sentiment analysis and sentence structure analysis in Korean, which is a typical agglutinative language with developed postpositions and endings. A morpheme can be defined as the smallest meaningful unit of a language, and a word consists of one or more morphemes. For example, for a word '예쁘고', the morphemes are '예쁘(= adjective)' and '고(=connective ending)'. Reflecting the significance of Korean morphemes, it seems reasonable to adopt the morphemes as a basic unit in Korean sentiment analysis. Therefore, in this study, we use 'morpheme vector' as an input to a deep learning model rather than 'word vector' which is mainly used in English text. The morpheme vector refers to a vector representation for the morpheme and can be derived by applying an existent word vector derivation mechanism to the sentences divided into constituent morphemes. By the way, here come some questions as follows. What is the desirable range of POS(Part-Of-Speech) tags when deriving morpheme vectors for improving the classification accuracy of a deep learning model? Is it proper to apply a typical word vector model which primarily relies on the form of words to Korean with a high homonym ratio? Will the text preprocessing such as correcting spelling or spacing errors affect the classification accuracy, especially when drawing morpheme vectors from Korean product reviews with a lot of grammatical mistakes and variations? We seek to find empirical answers to these fundamental issues, which may be encountered first when applying various deep learning models to Korean texts. As a starting point, we summarized these issues as three central research questions as follows. First, which is better effective, to use morpheme vectors from grammatically correct texts of other domain than the analysis target, or to use morpheme vectors from considerably ungrammatical texts of the same domain, as the initial input of a deep learning model? Second, what is an appropriate morpheme vector derivation method for Korean regarding the range of POS tags, homonym, text preprocessing, minimum frequency? Third, can we get a satisfactory level of classification accuracy when applying deep learning to Korean sentiment analysis? As an approach to these research questions, we generate various types of morpheme vectors reflecting the research questions and then compare the classification accuracy through a non-static CNN(Convolutional Neural Network) model taking in the morpheme vectors. As for training and test datasets, Naver Shopping's 17,260 cosmetics product reviews are used. To derive morpheme vectors, we use data from the same domain as the target one and data from other domain; Naver shopping's about 2 million cosmetics product reviews and 520,000 Naver News data arguably corresponding to Google's News data. The six primary sets of morpheme vectors constructed in this study differ in terms of the following three criteria. First, they come from two types of data source; Naver news of high grammatical correctness and Naver shopping's cosmetics product reviews of low grammatical correctness. Second, they are distinguished in the degree of data preprocessing, namely, only splitting sentences or up to additional spelling and spacing corrections after sentence separation. Third, they vary concerning the form of input fed into a word vector model; whether the morphemes themselves are entered into a word vector model or with their POS tags attached. The morpheme vectors further vary depending on the consideration range of POS tags, the minimum frequency of morphemes included, and the random initialization range. All morpheme vectors are derived through CBOW(Continuous Bag-Of-Words) model with the context window 5 and the vector dimension 300. It seems that utilizing the same domain text even with a lower degree of grammatical correctness, performing spelling and spacing corrections as well as sentence splitting, and incorporating morphemes of any POS tags including incomprehensible category lead to the better classification accuracy. The POS tag attachment, which is devised for the high proportion of homonyms in Korean, and the minimum frequency standard for the morpheme to be included seem not to have any definite influence on the classification accuracy.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.1
• /
• pp.25-36
• /
• 2008

Eye localization means localization of the center of the pupils, and is necessary for face recognition and related applications. Most of eye localization methods reported so far still need to be improved about robustness as well as precision for successful applications. In this paper, we propose a robust eye localization method using multi-scale Gabor feature vectors without big computational burden. The eye localization method using Gabor feature vectors is already employed in fuck as EBGM, but the method employed in EBGM is known not to be robust with respect to initial values, illumination, and pose, and may need extensive search range for achieving the required performance, which may cause big computational burden. The proposed method utilizes multi-scale approach. The proposed method first tries to localize eyes in the lower resolution face image by utilizing Gabor Jet similarity between Gabor feature vector at an estimated initial eye coordinates and the Gabor feature vectors in the eye model of the corresponding scale. Then the method localizes eyes in the next scale resolution face image in the same way but with initial eye points estimated from the eye coordinates localized in the lower resolution images. After repeating this process in the same way recursively, the proposed method funally localizes eyes in the original resolution face image. Also, the proposed method provides an effective illumination normalization to make the proposed multi-scale approach more robust to illumination, and additionally applies the illumination normalization technique in the preprocessing stage of the multi-scale approach so that the proposed method enhances the eye detection success rate. Experiment results verify that the proposed eye localization method improves the precision rate without causing big computational overhead compared to other eye localization methods reported in the previous researches and is robust to the variation of post: and illumination.

• Journal of Korea Multimedia Society
• /
• v.15 no.5
• /
• pp.595-605
• /
• 2012

Various methods for measuring strabismus have been developed and used in clinical diagnosis. However, most of them are based on the visual inspection by clinicians. For this reason, there is a high possibility of subjective evaluation in clinical decisions and they are only useful for cooperative patients. Therefore, the development of a more objective and reproducible method for measuring strabismus is needed. In this paper, we introduce a new software to complement the limitations of previous diagnostic methods. Firstly, we simply obtained facial images of patients and performed several preprocessing steps based on the spherical RGB color model with them. Then, the measurement of strabismus was performed automatically by using our 3D eye model and mathematical algorithm. To evaluate the validity of our software, we performed statistical correlation analysis of the results of the proposed method and the Krimsky test by two clinicians for ten patients. The coefficients of correlation for two clinicians were very high, 0.955 and 0.969, respectively. The coefficient of correlation between two clinicians also showed 0.968. We found a statistically significant correlation between two methods from our results. The newly developed software showed a possibility that it can be used as an alternative or effective assistant tool of previous diagnostic methods for strabismus.

• Journal of KIISE:Software and Applications
• /
• v.37 no.1
• /
• pp.29-38
• /
• 2010

We propose an interactive projection technique to display details of a large image in a high resolution and brightness by tracking a portable projector. A closed-loop based tracking method is presented to update the projected image while a user changes the position of the detail area by moving the portable projector. A marker is embedded in the large image to indicate the position to be occupied by the detail image projected by the portable projector. The marker is extracted in sequential images acquired by a camera attached to the portable projector. The marker position in the large display image is updated under a constraint that the center positions of marker and camera frame coincide in every camera frame. The image and projective transformation for warping are calculated using the marker position and shape in the camera frame. The marker's four corner points are determined by a four-step segmentation process which consists of camera image preprocessing based on HSI, edge extraction by Hough transformation, quadrangle test, and cross-ratio test. The interactive projection system implemented by the proposed method performs at about 24fps. In the user study, the overall feedback about the system usability was very high.

• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1449-1461
• /
• 2005

In this paper, we propose a vehicle recognition system based on the classification of vehicle identification mark and recognition of vehicle license plate. In the proposed algorithm, From the input vehicle image, we first simulate preprocessing procedures such as noise reduction, thinning etc., and detect vehicle identification mark and license plate region using the frequency distribution of intensity variation. And then, we classify extracted vehicle candidate region into identification mark, character and number of vehicle by using structural feature informations of vehicle. Lastly, we recognize vehicle informations with recognition of identification mark, character and number of vehicle using hybrid and vertical/horizontal pattern vector method. In the proposed algorithm, we used three properties of vehicle informations such as Independency property, discriminance property and frequency distribution of intensity variation property. In the vehicle images, identification mark is generally independent of the types of vehicle and vehicle identification mark. And also, the license plate region between character and background as well as horizontal/vertical intensity variations are more noticeable than other regions. To show the efficiency of the propofed algorithm, we tested it on 350 vehicle images and found that the propofed method shows good Performance regardless of irregular environment conditions as well as noise, size, and location of vehicles.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1440-1447
• /
• 2019

Currently, the prevalence of autism spectrum disorders in children is reported to be higher and shows various types of disorders. In particular, they are having difficulty in communication due to communication impairment in the area of social communication and need to be improved through training. Thus, this study proposes a method of acquiring voice information through a microphone mounted on a robot designed through preliminary research and using this information to make intelligent motions. An ANN(Artificial Neural Network) was used to classify the speech data into robot motions, and we tried to improve the accuracy by combining the Recurrent Neural Network based on Convolutional Neural Network. The preprocessing of input speech data was analyzed using MFCC(Mel-Frequency Cepstral Coefficient), and the motion of the robot was estimated using various data normalization and neural network optimization techniques. In addition, the designed ANN showed a high accuracy by conducting an experiment comparing the accuracy with the existing architecture and the method of human intervention. In order to design robot motions with higher accuracy in the future and to apply them in the treatment and education environment of children with autism.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.11
• /
• pp.91-99
• /
• 2010

Recently, recognizing user resource for personalized service has been needed in digital content service fields. Especially, to analyze user taste, recommend music and service music related information need recognition of user music file in case of online music service. Music related information service is offered through recognizing user music based on tag information. Recognition error has grown by weak points like changing and removing of tag information. Techniques of content based user music recognition with music signal itself are researched for solving upper problems. In this paper, we propose user music recognition on the internet by extracted feature from music signal. Features are extracted after suitable preprocessing for structure of content based user music recognition. Recognizing on music server consist of feature form are progressed with extracted feature. Through this, user music can be recognized independently of tag data. 600 music was collected and converted to each 5 music qualities for proving of proposed recognition. Converted 3000 experiment music on this method is used for recognition experiment on music server including 300,000 music. Average of recognition ratio was 85%. Weak points of tag based music recognition were overcome through proposed content based music recognition. Recognition performance of proposed method show a possibility that can be adapt to online music service in practice.

• Journal of Korea Game Society
• /
• v.7 no.1
• /
• pp.31-42
• /
• 2007

In the natural world, most materials such as skin, marble and cloth are translucent. Their appearance is smooth and soft compared with metals or mirrors. In this paper, we propose a new GPU based hierarchical rendering technique for translucent materials, based on the dipole diffusion approximation, at interactive rates. Information of incident light, position, normal, and irradiance, on the surfaces are stored into 2D textures by rendering from a primary light view. Huge numbers of pixel photons are clustered into quad-tree image pyramids. Each pixel, we select clusters (sets of photons), and then we approximate multiple subsurface scattering term with the clusters. We also introduce a novel hierarchical screen-space interpolation technique by exploiting spatial coherence with early-z culling on the GPU. We also build image pyramids of the screen using mipmap and pixel shader. Each pixel of the pyramids is stores position, normal and spatial similarity of children pixels. If a pixel's the similarity is high, we render the pixel and interpolate the pixel to multiple pixels. Result images show that our method can interactively render deformable translucent objects by approximating hundreds of thousand photons with only hundreds clusters without any preprocessing. We use an image-space approach for entire process on the GPU, thus our method is less dependent to scene complexity.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.4
• /
• pp.825-832
• /
• 2014

Previous works for detecting arrhythmia have mostly used nonlinear method such as artificial neural network, fuzzy theory, support vector machine to increase classification accuracy. Most methods require accurate detection of P-QRS-T point, higher computational cost and larger processing time. Even if some methods have the advantage in low complexity, but they generally suffer form low sensitivity. Also, it is difficult to detect PVC accurately because of the various QRS pattern by person's individual difference. Therefore it is necessary to design an efficient algorithm that classifies PVC based on QRS pattern in realtime and decreases computational cost by extracting minimal feature. In this paper, we propose PVC classification based on QRS pattern using QS interval and R wave amplitude. For this purpose, we detected R wave, RR interval, QRS pattern from noise-free ECG signal through the preprocessing method. Also, we classified PVC in realtime through QS interval and R wave amplitude. The performance of R wave detection, PVC classification is evaluated by using 9 record of MIT-BIH arrhythmia database that included over 30 PVC. The achieved scores indicate the average of 99.02% in R wave detection and the rate of 93.72% in PVC classification.