• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.73-85
• /
• 2013

Nowadays, in today's information society, the importance of the knowledge service using the information to creative value is getting higher day by day. In addition, depending on the development of IT technology, it is ease to collect and use information. Also, many companies actively use customer information to marketing in a variety of industries. Into the 21st century, companies have been actively using the culture arts to manage corporate image and marketing closely linked to their commercial interests. But, it is difficult that companies attract or maintain consumer's interest through their technology. For that reason, it is trend to perform cultural activities for tool of differentiation over many firms. Many firms used the customer's experience to new marketing strategy in order to effectively respond to competitive market. Accordingly, it is emerging rapidly that the necessity of personalized service to provide a new experience for people based on the personal profile information that contains the characteristics of the individual. Like this, personalized service using customer's individual profile information such as language, symbols, behavior, and emotions is very important today. Through this, we will be able to judge interaction between people and content and to maximize customer's experience and satisfaction. There are various relative works provide customer-centered service. Specially, emotion recognition research is emerging recently. Existing researches experienced emotion recognition using mostly bio-signal. Most of researches are voice and face studies that have great emotional changes. However, there are several difficulties to predict people's emotion caused by limitation of equipment and service environments. So, in this paper, we develop emotion prediction model based on vision-based interface to overcome existing limitations. Emotion recognition research based on people's gesture and posture has been processed by several researchers. This paper developed a model that recognizes people's emotional states through body gesture and posture using difference image method. And we found optimization validation model for four kinds of emotions' prediction. A proposed model purposed to automatically determine and predict 4 human emotions (Sadness, Surprise, Joy, and Disgust). To build up the model, event booth was installed in the KOCCA's lobby and we provided some proper stimulative movie to collect their body gesture and posture as the change of emotions. And then, we extracted body movements using difference image method. And we revised people data to build proposed model through neural network. The proposed model for emotion prediction used 3 type time-frame sets (20 frames, 30 frames, and 40 frames). And then, we adopted the model which has best performance compared with other models.' Before build three kinds of models, the entire 97 data set were divided into three data sets of learning, test, and validation set. The proposed model for emotion prediction was constructed using artificial neural network. In this paper, we used the back-propagation algorithm as a learning method, and set learning rate to 10%, momentum rate to 10%. The sigmoid function was used as the transform function. And we designed a three-layer perceptron neural network with one hidden layer and four output nodes. Based on the test data set, the learning for this research model was stopped when it reaches 50000 after reaching the minimum error in order to explore the point of learning. We finally processed each model's accuracy and found best model to predict each emotions. The result showed prediction accuracy 100% from sadness, and 96% from joy prediction in 20 frames set model. And 88% from surprise, and 98% from disgust in 30 frames set model. The findings of our research are expected to be useful to provide effective algorithm for personalized service in various industries such as advertisement, exhibition, performance, etc.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.141-154
• /
• 2019

Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.27-42
• /
• 2020

Traditional companies with offline stores were unable to secure large display space due to the problems of cost. This limitation inevitably allowed limited kinds of products to be displayed on the shelves, which resulted in consumers being deprived of the opportunity to experience various items. Taking advantage of the virtual space called the Internet, online shopping goes beyond the limits of limitations in physical space of offline shopping and is now able to display numerous products on web pages that can satisfy consumers with a variety of needs. Paradoxically, however, this can also cause consumers to experience the difficulty of comparing and evaluating too many alternatives in their purchase decision-making process. As an effort to address this side effect, various kinds of consumer's purchase decision support systems have been studied, such as keyword-based item search service and recommender systems. These systems can reduce search time for items, prevent consumer from leaving while browsing, and contribute to the seller's increased sales. Among those systems, recommender systems based on association rule mining techniques can effectively detect interrelated products from transaction data such as orders. The association between products obtained by statistical analysis provides clues to predicting how interested consumers will be in another product. However, since its algorithm is based on the number of transactions, products not sold enough so far in the early days of launch may not be included in the list of recommendations even though they are highly likely to be sold. Such missing items may not have sufficient opportunities to be exposed to consumers to record sufficient sales, and then fall into a vicious cycle of a vicious cycle of declining sales and omission in the recommendation list. This situation is an inevitable outcome in situations in which recommendations are made based on past transaction histories, rather than on determining potential future sales possibilities. This study started with the idea that reflecting the means by which this potential possibility can be identified indirectly would help to select highly recommended products. In the light of the fact that the attributes of a product affect the consumer's purchasing decisions, this study was conducted to reflect them in the recommender systems. In other words, consumers who visit a product page have shown interest in the attributes of the product and would be also interested in other products with the same attributes. On such assumption, based on these attributes, the recommender system can select recommended products that can show a higher acceptance rate. Given that a category is one of the main attributes of a product, it can be a good indicator of not only direct associations between two items but also potential associations that have yet to be revealed. Based on this idea, the study devised a recommender system that reflects not only associations between products but also categories. Through regression analysis, two kinds of associations were combined to form a model that could predict the hit rate of recommendation. To evaluate the performance of the proposed model, another regression model was also developed based only on associations between products. Comparative experiments were designed to be similar to the environment in which products are actually recommended in online shopping malls. First, the association rules for all possible combinations of antecedent and consequent items were generated from the order data. Then, hit rates for each of the associated rules were predicted from the support and confidence that are calculated by each of the models. The comparative experiments using order data collected from an online shopping mall show that the recommendation accuracy can be improved by further reflecting not only the association between products but also categories in the recommendation of related products. The proposed model showed a 2 to 3 percent improvement in hit rates compared to the existing model. From a practical point of view, it is expected to have a positive effect on improving consumers' purchasing satisfaction and increasing sellers' sales.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.57-78
• /
• 2020

With the development of the Internet, consumers have had an opportunity to check product information easily through E-Commerce. Product reviews used in the process of purchasing goods are based on user experience, allowing consumers to engage as producers of information as well as refer to information. This can be a way to increase the efficiency of purchasing decisions from the perspective of consumers, and from the seller's point of view, it can help develop products and strengthen their competitiveness. However, it takes a lot of time and effort to understand the overall assessment and assessment dimensions of the products that I think are important in reading the vast amount of product reviews offered by E-Commerce for the products consumers want to compare. This is because product reviews are unstructured information and it is difficult to read sentiment of reviews and assessment dimension immediately. For example, consumers who want to purchase a laptop would like to check the assessment of comparative products at each dimension, such as performance, weight, delivery, speed, and design. Therefore, in this paper, we would like to propose a method to automatically generate multi-dimensional product assessment scores in product reviews that we would like to compare. The methods presented in this study consist largely of two phases. One is the pre-preparation phase and the second is the individual product scoring phase. In the pre-preparation phase, a dimensioned classification model and a sentiment analysis model are created based on a review of the large category product group review. By combining word embedding and association analysis, the dimensioned classification model complements the limitation that word embedding methods for finding relevance between dimensions and words in existing studies see only the distance of words in sentences. Sentiment analysis models generate CNN models by organizing learning data tagged with positives and negatives on a phrase unit for accurate polarity detection. Through this, the individual product scoring phase applies the models pre-prepared for the phrase unit review. Multi-dimensional assessment scores can be obtained by aggregating them by assessment dimension according to the proportion of reviews organized like this, which are grouped among those that are judged to describe a specific dimension for each phrase. In the experiment of this paper, approximately 260,000 reviews of the large category product group are collected to form a dimensioned classification model and a sentiment analysis model. In addition, reviews of the laptops of S and L companies selling at E-Commerce are collected and used as experimental data, respectively. The dimensioned classification model classified individual product reviews broken down into phrases into six assessment dimensions and combined the existing word embedding method with an association analysis indicating frequency between words and dimensions. As a result of combining word embedding and association analysis, the accuracy of the model increased by 13.7%. The sentiment analysis models could be seen to closely analyze the assessment when they were taught in a phrase unit rather than in sentences. As a result, it was confirmed that the accuracy was 29.4% higher than the sentence-based model. Through this study, both sellers and consumers can expect efficient decision making in purchasing and product development, given that they can make multi-dimensional comparisons of products. In addition, text reviews, which are unstructured data, were transformed into objective values such as frequency and morpheme, and they were analysed together using word embedding and association analysis to improve the objectivity aspects of more precise multi-dimensional analysis and research. This will be an attractive analysis model in terms of not only enabling more effective service deployment during the evolving E-Commerce market and fierce competition, but also satisfying both customers.

• The Korean Journal of Nuclear Medicine
• /
• v.36 no.4
• /
• pp.244-254
• /
• 2002

Purpose: This study investigated the differences between technetium-99m ethyl cysteinate dimer (Tc-99m ECD) and technetium-99m hexamethylpropylene amine oxime (Tc-99m HMPAO) uptake in the normal brain by means of statistical parametric mapping (SPM) analysis. Materials and Methods: We retrospectively analyzed age and sex matched 53 cases of normal brain SPECT. Thirty-two cases were obtained with Tc-99m ECD and 21 cases with Tc-99m HMPAO. There were no abnormal findings on brain MRIs. All of the SPECT images were spatially transformed to standard space, smoothed and globally normalized. The differences between the Tc-99m ECD and Tc-99m HMPAO SPECT images were statistically analyzed using statistical parametric mapping (SPM'99) software. The differences bgetween the two groups were considered significant ant a threshold of corrected P values less than 0.05. Results: SPM analysis revealed significantly different uptakes of Tc-99m ECD and Tc-99m HMPAO in the normal brains. On the Tc-99m ECD SPECT images, relatively higher uptake was observed in the frontal, parietal and occipital lobes, in the basal ganglia and thalamus, and in the superior region of the cerebellum. On the Tc-99m HMPAO SPECT images, relatively higher uptakes was observed in subcortical areas of the frontal region, temporal lobe, and posterior portion of inferior cerebellum. Conclusion: Uptake of Tc-99m ECD and Tc-99m HMPO in the normallooking brain was significantly different on SPM analysis. The selective use of Tc-99m ECD of Tc-99m HMPAO in brain SPECT imaging appears especially valuable for the interpretation of cerebral perfusion. Further investigation is necessary to determine which tracer is more accurate for diagnosing different clinical conditions.

• Nuclear Medicine and Molecular Imaging
• /
• v.42 no.4
• /
• pp.301-306
• /
• 2008

Purpose: Recently, a number of patients needed total thyroidectomy and high dose radioiodine therapy (HD-RAI) get increased more. The aim of this study is to evaluate whether pathological staging (PS) and serum thyroglobulin (sTG) level could replace the diagnostic I-123 scan for the determination of therapeutic dose of HD-RAI in patients with differentiated thyroid cancer. Materials and Methods: Fifty eight patients (M:F=13;45, age $44.5{\pm}11.5\;yrs$) who underwent total thyroidectomy and central or regional lymph node dissection due to differentiated thyroid cancer were enrolled. Diagnostic scan of I-123 and sTG assay were also performed on off state of thyroid hormone. The therapeutic doses of I-131 (TD) were determined by the extent of uptakes on diagnostic I-123 scan as a gold standard. PS was graded by the criteria recommended in 6th edition of AJCC cancer staging manual except consideration of age. For comparison of the determination of therapeutic doses, PS and sTG were compared with the results of I-123 scan. Results: All patients were underwent HD-RAI. Among them, five patients (8.6%) were treated with 100 mCi of I-131, fourty three (74.1%) with 150 mCi, six (10.3%) with 180 mCi, three (5.2%) with 200 mCi, and one (1.7%) with 250 mCi, respectively. On the assessment of PS, average TDs were $154{\pm}25\;mCi$ in stage I (n=9), $175{\pm}50\;mCi$ in stage II (n=4), $149{\pm}21\;mCi$ in stage III (n=38), and $161{\pm}20\;mCi$ in stage IV (n=7). The statistical significance was not shown between PS and TD (p=0.169). Among fifty two patients who had available sTG, 25 patients (48.1%) having below 2 ng/mL of sTG were treated with $149{\pm}26\;mCi$ of I-131, 9 patients (17.3%) having $2{\leq}\;sTG\;<5\;ng/mL$ with $156{\pm}17\;mCi$, 5 patients (9.6%) having $5{\leq}\;sTG\;<10\;ng/mL$ with $156{\pm}13\;mCi$, 7 patients (13.5%) having $10{\leq}sTG\;<50\;ng/mL$ with $147{\pm}24\;mCi$, and 6 patients (11.5%) having above 50 ng/mL with $175{\pm}42\;mCi$. The statistical significance between sTG level and TD (p=0.252) was not shown. Conclusion: In conclusion, PS and sTG could not replace the determination of TD using I-123 scan for first HD-RAI in patients with differentiated thyroid cancer.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.24 no.1
• /
• pp.27-32
• /
• 2020

Purpose Glomerular Filtration Rate(GFR) is an important indicator for evaluating renal function and monitoring the progress of renal disease. Currently, the method of measuring GFR in clinical trials by using serum creatinine value and ^99mTc-DTPA(diethylenetriamine pentaacetic acid) renal dynamic scan is still useful. After the Gates method of formula was announced, when ^99mTc-DTPA Renal dynamic scan is taken, it is applied the GFR is measured using a gamma camera. The purpose of this paper is to measure the GFR by applying the Gates method of formula. It is according to effect activity and matrix size that is related in the GFR. Materials and Methods Data from 5 adult patients (patient age = 62 ± 5, 3 males, 2 females) who had been examined ^99mTc-DTPA Renal dynamic scan were analyzed. A dynamic image was obtained for 21 minutes after instantaneous injection of ^99mTc-DTPA 15 mCi into the patient's vein. To evaluate the glomerular filtration rate according to changes in activity and matrix size, total counts were measured after setting regions of interest in both kidneys and tissues in 2-3 minutes. The distance from detector to the table was maintained at 30cm, and the capacity of the pre-syringe (PR) was set to 15, 20, 25, 30 mCi, and each the capacity of post-syringe (PO) was 1, 5, 10, 15 mCi is set to evaluate the activity change. And then, each matrix size was changed to 32 × 32, 64 × 64, 128 × 128, 256 × 256, 512 × 512, and 1024 × 1024 to compare and to evaluate the values. Results As the activity increased in matrix size, the difference in GFR gradually decreased from 52.95% at the maximum to 16.67% at the minimum. The GFR value according to the change of matrix size was similar to 2.4%, 0.2%, 0.2% of difference when changing from 128 to 256, 256 to 512, and 512 to 1024, but 54.3% of difference when changing from 32 to 64 and 39.43% of difference when changing from 64 to 128. Finally, based on the presently used protocol, 256 × 256, PR 15 mCi and PO 1 mCi, the GFR value was the largest difference with 82% in PR 15 mCi and PO 1 mCi. conditions, and at the least difference is 0.2% in the conditions of PR 30 mCi and PO 15 mCi. Conclusion Through this paper, it was confirmed that when measuring the GFR using the gate method in the ^99mTc-DTPA renal dynamic scan. The GFR was affected by activity and matrix size changes. Therefore, it is considered that when taking the ^99mTc-DTPA renal dynamic scan, is should be careful by applying appropriate parameters when calculating GFR in the every hospital.

• Journal of Hospice and Palliative Care
• /
• v.6 no.1
• /
• pp.22-33
• /
• 2003

Purpose : The purpose of this study was to evaluate the present status of patients' satisfaction and the reasons for any satisfaction or dissatisfaction in cancer pain management Methods : A cross-sectional survey was used to obtain the feedback about pain management. The results of the survey were collected from 59 in- or out-patient who had cancer treatment at two of the teaching hospitals in Seoul from July, 2002 to November, 2002. The data was obtained by a structured questionnaire based on the American Cancer Society Patient Outcome Questionnaire(APS-POQ) and other previous research. The clinical information for all patients were compiled by reviewing their medical records. Resuts : 1) The subjects' mean score of the worst pain was 6.77, the average pain score was 3.80, and the pain score after management was 2.93 for the past 24 hours. The mean score of total pain interference was $25.03{\pm}12.82$. Many of the subjects had false beliefs about pain such as 'the experience of pain is a sign that the illness has gotten worse', 'pain medicine should be 'saved' in case the pain gets worse' and 'people get addicted to pain medicine easily'. 2) 66.1% of the subjects were properly medicated with analgesics. 33.9% of the subjects reported use of various methods in controlling pain other than the prescribed medication. Only 33.9% of the subjects had a chance to be educated about pain management by doctors or nurses. 3) The mean score of patients' satisfaction with pain management was $4.19{\pm}1.14$. 72.9% of the subjects answered 'satisfied' with pain management. The reasons for dissatisfaction were 'the pain was not relieved even after the pain management', 'I was not quickly and promptly treated when I complained of pain', 'doctors and nurses didn't pay much attention to my complaints of pain.', and 'there was no appropriate information given on the methods of administration, effect duration and side effects of pain medicine.' The reasons for satisfaction were: 'the pain was relieved after the pain management.', 'doctors and nurses quickly and promptly controlled my pain.', 'doctors and nurses paid enough attention to my complaints of pain.' and 'trust in my physician'. 4) In pain severity or pain interference, no significant difference was found between the satisfied group and dissatisfied group. On the belief 'good patients avoid talking about pain', a significant difference was found between the satisfied group and dissatisfied group. Conclusions : The patients' satisfaction with cancer pain management has increased over the years but still about 30% of patients reported to be 'not satisfied' for various reasons. The results of this study suggest that patients' education should be done to improve satisfaction in the pain management program.

• Journal of Food Hygiene and Safety
• /
• v.32 no.6
• /
• pp.470-476
• /
• 2017

The purpose of this study was to investigate and evaluate the safety of the grains, nut products, beans and oilseeds being sold in Gyeonggi province by analyzing mycotoxins. A multi-mycotoxins analysis method based on LC-MS/MS was validated and applied for the determination of eight mycotoxins, including aflatoxins ($B_1$, $B_2$, $G_1$ and $G_2$), fumonisins ($B_1$, $B_2$), zearalenone and ochratoxcin A in 134 samples. The limit of detection (LOD) and limit of quantitation (LOQ) for the eight mycotoxins ranged from 0.14 to $8.25{\mu}g/kg$ and from 1.08 to $7.21{\mu}g/kg$, respectively. Recovery rates of mycotoxins were determined in the range of 61.1 to 97.5% with RSD of 1.0~14.5% (n=3). Fumonisin $B_1$, $B_2$, zearalenone, and ochratoxin A were detected in 22 samples, indicating that 27% of grains, 12.5% of beans and 11.8% of oilseeds were contaminated. Fumonisin and zearalenone were detected simultaneously in 2 adlays and 3 sorghums. Fumonisin $B_1$ and $B_2$ were detected simultaneously in most samples whereas fumonisin $B_1$ was detected in 1 adlay, 1 millet and 1 sesame sample. The average detected amount of fumonisin was $49.3{\mu}g/kg$ and $10.1{\mu}g/kg$ for grains and oilseeds, respectively. The average detected amount of zearalenone was $1.9{\mu}g/kg$ and $1.5{\mu}g/kg$ for grains and beans, respectively. In addition, the average amount of ochratoxin A was $0.08{\mu}g/kg$ for grains. The calculated exposure amounts of fumonisin, zeralenone and ochratoxin A for grains, beans and oilseeds were below the PMTDI/PTWI.

• Journal of Nutrition and Health
• /
• v.50 no.3
• /
• pp.270-283
• /
• 2017

Purpose: This study was conducted to examine coffee consumption behaviors, dietary habits, and nutrient intakes by coffee intake amount among university students. Methods: Questionnaires were distributed to 300 university students randomly selected in Gongju. Dietary survey was administered during two weekdays by the food record method. Results: Subjects were divided into three groups: NCG (non-coffee group), LCG (low coffee group, 1~2 cups/d), and HCG (high coffee group, 3 cups/d) by coffee intake amount and subjects' distribution. Coffee intake frequency was significantly greater in the HCG compared to the LCG (p < 0.001). The HCG was more likely to intake dripped coffee with or without milk and/or sugar than the LCG (p < 0.05). More than 80% of coffee drinkers chose their favorite coffee or accompanying snacks regardless of energy content. More than 75% of coffee takers did not eat accompanying snacks instead of meals, and the HCG ate them more frequently than LCG (p < 0.05). Breakfast skipping rate was high while vegetable and fruit intakes were very low in most subjects. Subjects who drank carbonated drinks, sweet beverages, or alcohol were significantly greater in number in the LCG and HCG than in the NCG (p < 0.01). Energy intakes from coffee were $0.88{\pm}5.62kcal/d$ and $7.07{\pm}16.93kcal/d$ for the LCG and HCG. For total subjects, daily mean dietary energy intake was low at less than 72% of estimated energy requirement. Levels of vitamin C and calcium were lower than the estimated average requirements while that of vitamin D was low (24~34% of adequate intake). There was no difference in nutrient intakes by coffee intake amount, except protein, vitamin A, and niacin. Conclusion: Coffee intake amount did not affect dietary nutrient intakes. Dietary habits were poor,and most nutrient intakes were lower than recommend levels. High intakes of coffee seemed to be related with high consumption of sweet beverages and alcohol. Therefore, it is necessary to improve nutritional intakes and encourage proper water intake habits, including coffee intake, for improved nutritional status of subjects.