• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.63-83
• /
• 2019

Investors prefer to look for trading points based on the graph shown in the chart rather than complex analysis, such as corporate intrinsic value analysis and technical auxiliary index analysis. However, the pattern analysis technique is difficult and computerized less than the needs of users. In recent years, there have been many cases of studying stock price patterns using various machine learning techniques including neural networks in the field of artificial intelligence(AI). In particular, the development of IT technology has made it easier to analyze a huge number of chart data to find patterns that can predict stock prices. Although short-term forecasting power of prices has increased in terms of performance so far, long-term forecasting power is limited and is used in short-term trading rather than long-term investment. Other studies have focused on mechanically and accurately identifying patterns that were not recognized by past technology, but it can be vulnerable in practical areas because it is a separate matter whether the patterns found are suitable for trading. When they find a meaningful pattern, they find a point that matches the pattern. They then measure their performance after n days, assuming that they have bought at that point in time. Since this approach is to calculate virtual revenues, there can be many disparities with reality. The existing research method tries to find a pattern with stock price prediction power, but this study proposes to define the patterns first and to trade when the pattern with high success probability appears. The M & W wave pattern published by Merrill(1980) is simple because we can distinguish it by five turning points. Despite the report that some patterns have price predictability, there were no performance reports used in the actual market. The simplicity of a pattern consisting of five turning points has the advantage of reducing the cost of increasing pattern recognition accuracy. In this study, 16 patterns of up conversion and 16 patterns of down conversion are reclassified into ten groups so that they can be easily implemented by the system. Only one pattern with high success rate per group is selected for trading. Patterns that had a high probability of success in the past are likely to succeed in the future. So we trade when such a pattern occurs. It is a real situation because it is measured assuming that both the buy and sell have been executed. We tested three ways to calculate the turning point. The first method, the minimum change rate zig-zag method, removes price movements below a certain percentage and calculates the vertex. In the second method, high-low line zig-zag, the high price that meets the n-day high price line is calculated at the peak price, and the low price that meets the n-day low price line is calculated at the valley price. In the third method, the swing wave method, the high price in the center higher than n high prices on the left and right is calculated as the peak price. If the central low price is lower than the n low price on the left and right, it is calculated as valley price. The swing wave method was superior to the other methods in the test results. It is interpreted that the transaction after checking the completion of the pattern is more effective than the transaction in the unfinished state of the pattern. Genetic algorithms(GA) were the most suitable solution, although it was virtually impossible to find patterns with high success rates because the number of cases was too large in this simulation. We also performed the simulation using the Walk-forward Analysis(WFA) method, which tests the test section and the application section separately. So we were able to respond appropriately to market changes. In this study, we optimize the stock portfolio because there is a risk of over-optimized if we implement the variable optimality for each individual stock. Therefore, we selected the number of constituent stocks as 20 to increase the effect of diversified investment while avoiding optimization. We tested the KOSPI market by dividing it into six categories. In the results, the portfolio of small cap stock was the most successful and the high vol stock portfolio was the second best. This shows that patterns need to have some price volatility in order for patterns to be shaped, but volatility is not the best.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.85-109
• /
• 2018

Recommender system recommends the items expected to be purchased by a customer in the future according to his or her previous purchase behaviors. It has been served as a tool for realizing one-to-one personalization for an e-commerce service company. Traditional recommender systems, especially the recommender systems based on collaborative filtering (CF), which is the most popular recommendation algorithm in both academy and industry, are designed to generate the items list for recommendation by using 'overall rating' - a single criterion. However, it has critical limitations in understanding the customers' preferences in detail. Recently, to mitigate these limitations, some leading e-commerce companies have begun to get feedback from their customers in a form of 'multicritera ratings'. Multicriteria ratings enable the companies to understand their customers' preferences from the multidimensional viewpoints. Moreover, it is easy to handle and analyze the multidimensional ratings because they are quantitative. But, the recommendation using multicritera ratings also has limitation that it may omit detail information on a user's preference because it only considers three-to-five predetermined criteria in most cases. Under this background, this study proposes a novel hybrid recommendation system, which selectively uses the results from 'traditional CF' and 'CF using multicriteria ratings'. Our proposed system is based on the premise that some people have holistic preference scheme, whereas others have composite preference scheme. Thus, our system is designed to use traditional CF using overall rating for the users with holistic preference, and to use CF using multicriteria ratings for the users with composite preference. To validate the usefulness of the proposed system, we applied it to a real-world dataset regarding the recommendation for POI (point-of-interests). Providing personalized POI recommendation is getting more attentions as the popularity of the location-based services such as Yelp and Foursquare increases. The dataset was collected from university students via a Web-based online survey system. Using the survey system, we collected the overall ratings as well as the ratings for each criterion for 48 POIs that are located near K university in Seoul, South Korea. The criteria include 'food or taste', 'price' and 'service or mood'. As a result, we obtain 2,878 valid ratings from 112 users. Among 48 items, 38 items (80%) are used as training dataset, and the remaining 10 items (20%) are used as validation dataset. To examine the effectiveness of the proposed system (i.e. hybrid selective model), we compared its performance to the performances of two comparison models - the traditional CF and the CF with multicriteria ratings. The performances of recommender systems were evaluated by using two metrics - average MAE(mean absolute error) and precision-in-top-N. Precision-in-top-N represents the percentage of truly high overall ratings among those that the model predicted would be the N most relevant items for each user. The experimental system was developed using Microsoft Visual Basic for Applications (VBA). The experimental results showed that our proposed system (avg. MAE = 0.584) outperformed traditional CF (avg. MAE = 0.591) as well as multicriteria CF (avg. AVE = 0.608). We also found that multicriteria CF showed worse performance compared to traditional CF in our data set, which is contradictory to the results in the most previous studies. This result supports the premise of our study that people have two different types of preference schemes - holistic and composite. Besides MAE, the proposed system outperformed all the comparison models in precision-in-top-3, precision-in-top-5, and precision-in-top-7. The results from the paired samples t-test presented that our proposed system outperformed traditional CF with 10% statistical significance level, and multicriteria CF with 1% statistical significance level from the perspective of average MAE. The proposed system sheds light on how to understand and utilize user's preference schemes in recommender systems domain.

• Journal of Intelligence and Information Systems
• /
• v.16 no.3
• /
• pp.77-97
• /
• 2010

Market timing is an investment strategy which is used for obtaining excessive return from financial market. In general, detection of market timing means determining when to buy and sell to get excess return from trading. In many market timing systems, trading rules have been used as an engine to generate signals for trade. On the other hand, some researchers proposed the rough set analysis as a proper tool for market timing because it does not generate a signal for trade when the pattern of the market is uncertain by using the control function. The data for the rough set analysis should be discretized of numeric value because the rough set only accepts categorical data for analysis. Discretization searches for proper "cuts" for numeric data that determine intervals. All values that lie within each interval are transformed into same value. In general, there are four methods for data discretization in rough set analysis including equal frequency scaling, expert's knowledge-based discretization, minimum entropy scaling, and na$\ddot{i}$ve and Boolean reasoning-based discretization. Equal frequency scaling fixes a number of intervals and examines the histogram of each variable, then determines cuts so that approximately the same number of samples fall into each of the intervals. Expert's knowledge-based discretization determines cuts according to knowledge of domain experts through literature review or interview with experts. Minimum entropy scaling implements the algorithm based on recursively partitioning the value set of each variable so that a local measure of entropy is optimized. Na$\ddot{i}$ve and Booleanreasoning-based discretization searches categorical values by using Na$\ddot{i}$ve scaling the data, then finds the optimized dicretization thresholds through Boolean reasoning. Although the rough set analysis is promising for market timing, there is little research on the impact of the various data discretization methods on performance from trading using the rough set analysis. In this study, we compare stock market timing models using rough set analysis with various data discretization methods. The research data used in this study are the KOSPI 200 from May 1996 to October 1998. KOSPI 200 is the underlying index of the KOSPI 200 futures which is the first derivative instrument in the Korean stock market. The KOSPI 200 is a market value weighted index which consists of 200 stocks selected by criteria on liquidity and their status in corresponding industry including manufacturing, construction, communication, electricity and gas, distribution and services, and financing. The total number of samples is 660 trading days. In addition, this study uses popular technical indicators as independent variables. The experimental results show that the most profitable method for the training sample is the na$\ddot{i}$ve and Boolean reasoning but the expert's knowledge-based discretization is the most profitable method for the validation sample. In addition, the expert's knowledge-based discretization produced robust performance for both of training and validation sample. We also compared rough set analysis and decision tree. This study experimented C4.5 for the comparison purpose. The results show that rough set analysis with expert's knowledge-based discretization produced more profitable rules than C4.5.

• 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.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• 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.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.219-240
• /
• 2018

Sentiment analysis, which is one of the text mining techniques, is a method for extracting subjective content embedded in text documents. Recently, the sentiment analysis methods have been widely used in many fields. As good examples, data-driven surveys are based on analyzing the subjectivity of text data posted by users and market researches are conducted by analyzing users' review posts to quantify users' reputation on a target product. The basic method of sentiment analysis is to use sentiment dictionary (or lexicon), a list of sentiment vocabularies with positive, neutral, or negative semantics. In general, the meaning of many sentiment words is likely to be different across domains. For example, a sentiment word, 'sad' indicates negative meaning in many fields but a movie. In order to perform accurate sentiment analysis, we need to build the sentiment dictionary for a given domain. However, such a method of building the sentiment lexicon is time-consuming and various sentiment vocabularies are not included without the use of general-purpose sentiment lexicon. In order to address this problem, several studies have been carried out to construct the sentiment lexicon suitable for a specific domain based on 'OPEN HANGUL' and 'SentiWordNet', which are general-purpose sentiment lexicons. However, OPEN HANGUL is no longer being serviced and SentiWordNet does not work well because of language difference in the process of converting Korean word into English word. There are restrictions on the use of such general-purpose sentiment lexicons as seed data for building the sentiment lexicon for a specific domain. In this article, we construct 'KNU Korean Sentiment Lexicon (KNU-KSL)', a new general-purpose Korean sentiment dictionary that is more advanced than existing general-purpose lexicons. The proposed dictionary, which is a list of domain-independent sentiment words such as 'thank you', 'worthy', and 'impressed', is built to quickly construct the sentiment dictionary for a target domain. Especially, it constructs sentiment vocabularies by analyzing the glosses contained in Standard Korean Language Dictionary (SKLD) by the following procedures: First, we propose a sentiment classification model based on Bidirectional Long Short-Term Memory (Bi-LSTM). Second, the proposed deep learning model automatically classifies each of glosses to either positive or negative meaning. Third, positive words and phrases are extracted from the glosses classified as positive meaning, while negative words and phrases are extracted from the glosses classified as negative meaning. Our experimental results show that the average accuracy of the proposed sentiment classification model is up to 89.45%. In addition, the sentiment dictionary is more extended using various external sources including SentiWordNet, SenticNet, Emotional Verbs, and Sentiment Lexicon 0603. Furthermore, we add sentiment information about frequently used coined words and emoticons that are used mainly on the Web. The KNU-KSL contains a total of 14,843 sentiment vocabularies, each of which is one of 1-grams, 2-grams, phrases, and sentence patterns. Unlike existing sentiment dictionaries, it is composed of words that are not affected by particular domains. The recent trend on sentiment analysis is to use deep learning technique without sentiment dictionaries. The importance of developing sentiment dictionaries is declined gradually. However, one of recent studies shows that the words in the sentiment dictionary can be used as features of deep learning models, resulting in the sentiment analysis performed with higher accuracy (Teng, Z., 2016). This result indicates that the sentiment dictionary is used not only for sentiment analysis but also as features of deep learning models for improving accuracy. The proposed dictionary can be used as a basic data for constructing the sentiment lexicon of a particular domain and as features of deep learning models. It is also useful to automatically and quickly build large training sets for deep learning models.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• Korean Journal of Cognitive Science
• /
• v.23 no.2
• /
• pp.133-164
• /
• 2012

Paul Churchland(1989) suggests the theory of representation from the results of cognitive biology and connectionist AI studies. According to the theory, our representations of the diverse phenomena in the world can be represented as the positions of phase state spaces with the actions of the neurons or of the assembly of neurons. He insists connectionist AI neural networks can have the semantical category systems to recognize the world. But Fodor and Lepore(1996) don't look the perspective bright. From their points of view, the Churchland's theory of representation stands on the base of Quine's holism, and the network semantics cannot explain how the criteria of semantical content similarity could be possible, and so cannot the theory. This thesis aims to excavate which one is the better between the perspective of the theory and the one of Fodor and Lepore's. From my understandings of state space theory of representation, artificial nets can coordinates the criteria of contents similarity by the learning algorithm. On the basis of these, I can see that Fodor and Lepore's points cannot penetrate the Churchlands' theory. From the view point of the theory, we can see how the future's artificial systems can have the conceptual systems recognizing the world. Therefore we can have the perspectives what cognitive scientists have to focus on.