• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.185-205
• /
• 2023

As the tourism industry recovers post the COVID-19 pandemic, an increasing number of tourists are utilizing various platforms to leave reviews. However, amidst the vast amount of data, finding useful information remains challenging, often leading to time and cost inefficiencies in selecting travel destinations. Despite ongoing research, there are limitations due to the absence of ratings or the presence of different rating formats across platforms. Moreover, inconsistencies between ratings and the content of reviews pose challenges in developing recommendation models. To address these issues, this study utilized 7,104 reviews of tourist spots in Jeju Island to develop a specialized satisfaction index for Jeju tourist attractions and employed this index to construct a 'Rating Prediction Model.' To validate the model's performance, we predicted the ratings of 700 experimental data points using both the developed model and an LSTM approach. The proposed model demonstrated superior performance with a weighted accuracy of 73.87%, which is approximately 4.67% higher than that of the LSTM. The results of this study are expected to resolve the discrepancies between ratings and review contents, standardize ratings in reviews without ratings or in various formats, and provide reliable rating indicators applicable across all areas of travel in different domains.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.525-531
• /
• 2009

Data mining is the process of extracting hidden patterns from collected data. At this time, for collected data which take important role as the basic information for prediction and recommendation, the process to discriminate incorrect data in order to enhance the performance of analysis result, is needed. The existing methods to discriminate unexpected data from collected data, mainly relies on methods which are based on statistics or simple distance between data. However, for these methods, the problematic point that even meaningful data could be excluded from analysis due that the environment and characteristic of the relevant data are not considered, exists. This study proposes a method to endow human heuristic knowledge with weight value through the comparison between collected data and human heuristic knowledge, and to use the value for creating a decision tree. The data discrimination by the method proposed is more credible as human knowledge is reflected in the created tree. The validity of the proposed method is verified through an experiment.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.11 no.2
• /
• pp.91-98
• /
• 2017

Background & Object: Basic daily activity screening tool such as the Modified Barthel Index (MBI) has been used commonly in rehabilitation clinic and community based rehabilitation setting. Previous studies have shown the significant relations between the level of daily activities and driving ability on stroke or elderly people. However, there is a lack of studies to investigate the usefulness of MBI on prediction of driving ability for stroke patient. This study was to predict driving abilities of stroke survivor using Korean version Modified Barthel Index (K-MBI). Methods: A sample of 48 patients with stroke in rehabilitation hospital was recruited. All participants were tested level of basic daily activities using K-MBI. The driving ability of participants was tested using virtual reality driving simulator. The predictive validity was calculated of the K-MBI among pass or fail group of driving simulator test using receiver operating characteristics curves. Results: The cut-off score of >86.5 on the K-MBI is proper sensitivity to predict on driving performance ability. Conclusion: This pilot result offers clinical reference to therapists and caregivers for reasoning on driving recommendation period during rehabilitation stage of stroke survivors. Further studies need to identify prediction using real on-road test in a large population group.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.79-96
• /
• 2012

As the pace of competition dramatically accelerates and the complexity of change grows, a variety of research have been conducted to improve firms' short-term performance and to enhance firms' long-term survival. In particular, researchers and practitioners have paid their attention to identify promising technologies that lead competitive advantage to a firm. Discovery of promising technology depends on how a firm evaluates the value of technologies, thus many evaluating methods have been proposed. Experts' opinion based approaches have been widely accepted to predict the value of technologies. Whereas this approach provides in-depth analysis and ensures validity of analysis results, it is usually cost-and time-ineffective and is limited to qualitative evaluation. Considerable studies attempt to forecast the value of technology by using patent information to overcome the limitation of experts' opinion based approach. Patent based technology evaluation has served as a valuable assessment approach of the technological forecasting because it contains a full and practical description of technology with uniform structure. Furthermore, it provides information that is not divulged in any other sources. Although patent information based approach has contributed to our understanding of prediction of promising technologies, it has some limitations because prediction has been made based on the past patent information, and the interpretations of patent analyses are not consistent. In order to fill this gap, this study proposes a technology forecasting methodology by integrating patent information approach and artificial intelligence method. The methodology consists of three modules : evaluation of technologies promising, implementation of technologies value prediction model, and recommendation of promising technologies. In the first module, technologies promising is evaluated from three different and complementary dimensions; impact, fusion, and diffusion perspectives. The impact of technologies refers to their influence on future technologies development and improvement, and is also clearly associated with their monetary value. The fusion of technologies denotes the extent to which a technology fuses different technologies, and represents the breadth of search underlying the technology. The fusion of technologies can be calculated based on technology or patent, thus this study measures two types of fusion index; fusion index per technology and fusion index per patent. Finally, the diffusion of technologies denotes their degree of applicability across scientific and technological fields. In the same vein, diffusion index per technology and diffusion index per patent are considered respectively. In the second module, technologies value prediction model is implemented using artificial intelligence method. This studies use the values of five indexes (i.e., impact index, fusion index per technology, fusion index per patent, diffusion index per technology and diffusion index per patent) at different time (e.g., t-n, t-n-1, t-n-2, ${\cdots}$) as input variables. The out variables are values of five indexes at time t, which is used for learning. The learning method adopted in this study is backpropagation algorithm. In the third module, this study recommends final promising technologies based on analytic hierarchy process. AHP provides relative importance of each index, leading to final promising index for technology. Applicability of the proposed methodology is tested by using U.S. patents in international patent class G06F (i.e., electronic digital data processing) from 2000 to 2008. The results show that mean absolute error value for prediction produced by the proposed methodology is lower than the value produced by multiple regression analysis in cases of fusion indexes. However, mean absolute error value of the proposed methodology is slightly higher than the value of multiple regression analysis. These unexpected results may be explained, in part, by small number of patents. Since this study only uses patent data in class G06F, number of sample patent data is relatively small, leading to incomplete learning to satisfy complex artificial intelligence structure. In addition, fusion index per technology and impact index are found to be important criteria to predict promising technology. This study attempts to extend the existing knowledge by proposing a new methodology for prediction technology value by integrating patent information analysis and artificial intelligence network. It helps managers who want to technology develop planning and policy maker who want to implement technology policy by providing quantitative prediction methodology. In addition, this study could help other researchers by proving a deeper understanding of the complex technological forecasting field.

• Journal of Intelligence and Information Systems
• /
• v.19 no.3
• /
• pp.57-71
• /
• 2013

Over a billion people in the world generate new news minute by minute. People forecasts some news but most news are from unexpected events such as natural disasters, accidents, crimes. People spend much time to watch a huge amount of news delivered from many media because they want to understand what is happening now, to predict what might happen in the near future, and to share and discuss on the news. People make better daily decisions through watching and obtaining useful information from news they saw. However, it is difficult that people choose news suitable to them and obtain useful information from the news because there are so many news media such as portal sites, broadcasters, and most news articles consist of gossipy news and breaking news. User interest changes over time and many people have no interest in outdated news. From this fact, applying users' recent interest to personalized news service is also required in news service. It means that personalized news service should dynamically manage user profiles. In this paper, a content-based news recommendation system is proposed to provide the personalized news service. For a personalized service, user's personal information is requisitely required. Social network service is used to extract user information for personalization service. The proposed system constructs dynamic user profile based on recent user information of Facebook, which is one of social network services. User information contains personal information, recent articles, and Facebook Page information. Facebook Pages are used for businesses, organizations and brands to share their contents and connect with people. Facebook users can add Facebook Page to specify their interest in the Page. The proposed system uses this Page information to create user profile, and to match user preferences to news topics. However, some Pages are not directly matched to news topic because Page deals with individual objects and do not provide topic information suitable to news. Freebase, which is a large collaborative database of well-known people, places, things, is used to match Page to news topic by using hierarchy information of its objects. By using recent Page information and articles of Facebook users, the proposed systems can own dynamic user profile. The generated user profile is used to measure user preferences on news. To generate news profile, news category predefined by news media is used and keywords of news articles are extracted after analysis of news contents including title, category, and scripts. TF-IDF technique, which reflects how important a word is to a document in a corpus, is used to identify keywords of each news article. For user profile and news profile, same format is used to efficiently measure similarity between user preferences and news. The proposed system calculates all similarity values between user profiles and news profiles. Existing methods of similarity calculation in vector space model do not cover synonym, hypernym and hyponym because they only handle given words in vector space model. The proposed system applies WordNet to similarity calculation to overcome the limitation. Top-N news articles, which have high similarity value for a target user, are recommended to the user. To evaluate the proposed news recommendation system, user profiles are generated using Facebook account with participants consent, and we implement a Web crawler to extract news information from PBS, which is non-profit public broadcasting television network in the United States, and construct news profiles. We compare the performance of the proposed method with that of benchmark algorithms. One is a traditional method based on TF-IDF. Another is 6Sub-Vectors method that divides the points to get keywords into six parts. Experimental results demonstrate that the proposed system provide useful news to users by applying user's social network information and WordNet functions, in terms of prediction error of recommended news.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.39-54
• /
• 2013

Recent explosive increase of electronic commerce provides many advantageous purchase opportunities to customers. In this situation, customers who do not have enough knowledge about their purchases, may accept product recommendations. Product recommender systems automatically reflect user's preference and provide recommendation list to the users. Thus, product recommender system in online shopping store has been known as one of the most popular tools for one-to-one marketing. However, recommender systems which do not properly reflect user's preference cause user's disappointment and waste of time. In this study, we propose a novel recommender system which uses data mining and multi-model ensemble techniques to enhance the recommendation performance through reflecting the precise user's preference. The research data is collected from the real-world online shopping store, which deals products from famous art galleries and museums in Korea. The data initially contain 5759 transaction data, but finally remain 3167 transaction data after deletion of null data. In this study, we transform the categorical variables into dummy variables and exclude outlier data. The proposed model consists of two steps. The first step predicts customers who have high likelihood to purchase products in the online shopping store. In this step, we first use logistic regression, decision trees, and artificial neural networks to predict customers who have high likelihood to purchase products in each product group. We perform above data mining techniques using SAS E-Miner software. In this study, we partition datasets into two sets as modeling and validation sets for the logistic regression and decision trees. We also partition datasets into three sets as training, test, and validation sets for the artificial neural network model. The validation dataset is equal for the all experiments. Then we composite the results of each predictor using the multi-model ensemble techniques such as bagging and bumping. Bagging is the abbreviation of "Bootstrap Aggregation" and it composite outputs from several machine learning techniques for raising the performance and stability of prediction or classification. This technique is special form of the averaging method. Bumping is the abbreviation of "Bootstrap Umbrella of Model Parameter," and it only considers the model which has the lowest error value. The results show that bumping outperforms bagging and the other predictors except for "Poster" product group. For the "Poster" product group, artificial neural network model performs better than the other models. In the second step, we use the market basket analysis to extract association rules for co-purchased products. We can extract thirty one association rules according to values of Lift, Support, and Confidence measure. We set the minimum transaction frequency to support associations as 5%, maximum number of items in an association as 4, and minimum confidence for rule generation as 10%. This study also excludes the extracted association rules below 1 of lift value. We finally get fifteen association rules by excluding duplicate rules. Among the fifteen association rules, eleven rules contain association between products in "Office Supplies" product group, one rules include the association between "Office Supplies" and "Fashion" product groups, and other three rules contain association between "Office Supplies" and "Home Decoration" product groups. Finally, the proposed product recommender systems provides list of recommendations to the proper customers. We test the usability of the proposed system by using prototype and real-world transaction and profile data. For this end, we construct the prototype system by using the ASP, Java Script and Microsoft Access. In addition, we survey about user satisfaction for the recommended product list from the proposed system and the randomly selected product lists. The participants for the survey are 173 persons who use MSN Messenger, Daum Caf$\acute{e}$, and P2P services. We evaluate the user satisfaction using five-scale Likert measure. This study also performs "Paired Sample T-test" for the results of the survey. The results show that the proposed model outperforms the random selection model with 1% statistical significance level. It means that the users satisfied the recommended product list significantly. The results also show that the proposed system may be useful in real-world online shopping store.

• Journal of Crop Science and Biotechnology
• /
• v.11 no.2
• /
• pp.141-150
• /
• 2008

The response of grain yield(GY) and milled-rice protein content(PC) to crop growth status and nitrogen(N) rates at panicle initiation stage(PIS) is critical information for prescribing topdress N rate at PIS(Npi) for target GY and PC. Three split-split-plot experiments including various N treatments and rice cultivars were conducted in Experimental Farm, Seoul National University, Korea in 2003-2005. Shoot N density(SND, g N in shoot $m^{-2}$) and canopy reflectance were measured before N application at PIS, and GY, PC, and SND were measured at harvest. Data from the first two years(2003-2004) were used for calibrating the predictive models for GY, PC, and SND accumulated from PIS to harvest using SND at PIS and Npi by multiple stepwise regression. After that the calibrated models were used for calculating N requirement at PIS for each of nine plots based on the target PC of 6.8% and the values of SND at PIS that was estimated by canopy reflectance method in the 2005 experiment. The result showed that SND at PIS in combination with Npi were successful to predict GY, PC, and SND from PIS to harvest in the calibration dataset with the coefficients of determination ($R^2$) of 0.87, 0.73, and 0.82 and the relative errors in prediction(REP, %) of 5.5, 4.3, and 21.1%, respectively. In general, the calibrated model equations showed a little lower performance in calculating GY, PC, and SND in the validation dataset(data from 2005) but REP ranging from 3.3% for PC and 13.9% for SND accumulated from PIS to harvest was acceptable. Nitrogen rate prescription treatment(PRT) for the target PC of 6.8% reduced the coefficient of variation in PC from 4.6% in the fixed rate treatment(FRT, 3.6g N $m^{-2}$) to 2.4% in PRT and the average PC of PRT was 6.78%, being very close to the target PC of 6.8%. In addition, PRT increased GY by 42.1 $gm^{-2}$ while Npi increased by 0.63 $gm^{-2}$ compared to the FRT, resulting in high agronomic N-use efficiency of 68.8 kg grain from additional kg N. The high agronomic N-use efficiency might have resulted from the higher response of grain yield to the applied N in the prescribed N rate treatment because N rate was prescribed based on the crop growth and N status of each plot.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.467-477
• /
• 2004

While recommender systems were used by a few E-commerce sites former days, they are now becoming serious business tools that are re-shaping the world of I-commerce. And collaborative filtering has been a very successful recommendation technique in both research and practice. But there are two problems in personalized recommender systems, it is First-Rating problem and Sparsity problem. In this paper, we solve these problems using the associative relation clustering and “Lift” of association rules. We produce “Lift” between items using user's rating data. And we apply Threshold by -cut to the association between items. To make an efficiency of associative relation cluster higher, we use not only the existing Hypergraph Clique Clustering algorithm but also the suggested Split Cluster method. If the cluster is completed, we calculate a similarity iten in each inner cluster. And the index is saved in the database for the fast access. We apply the creating index to predict the preference for new items. To estimate the Performance, the suggested method is compared with existing collaborative filtering techniques. As a result, the proposed method is efficient for improving the accuracy of prediction through solving problems of existing collaborative filtering techniques.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.95-108
• /
• 2011

TThis is a study of the personalization method that intelligently adapts the level of clustering considering purchasing index of a customer. In the e-biz era, many companies gather customers' demographic and transactional information such as age, gender, purchasing date and product category. They use this information to predict customer's preferences or purchasing patterns so that they can provide more customized services to their customers. The previous Customer-Segmentation method provides customized services for each customer group. This method clusters a whole customer set into different groups based on their similarity and builds predictive models for the resulting groups. Thus, it can manage the number of predictive models and also provide more data for the customers who do not have enough data to build a good predictive model by using the data of other similar customers. However, this method often fails to provide highly personalized services to each customer, which is especially important to VIP customers. Furthermore, it clusters the customers who already have a considerable amount of data as well as the customers who only have small amount of data, which causes to increase computational cost unnecessarily without significant performance improvement. The other conventional method called 1-to-1 method provides more customized services than the Customer-Segmentation method for each individual customer since the predictive model are built using only the data for the individual customer. This method not only provides highly personalized services but also builds a relatively simple and less costly model that satisfies with each customer. However, the 1-to-1 method has a limitation that it does not produce a good predictive model when a customer has only a few numbers of data. In other words, if a customer has insufficient number of transactional data then the performance rate of this method deteriorate. In order to overcome the limitations of these two conventional methods, we suggested the new method called Intelligent Customer Segmentation method that provides adaptive personalized services according to the customer's purchasing index. The suggested method clusters customers according to their purchasing index, so that the prediction for the less purchasing customers are based on the data in more intensively clustered groups, and for the VIP customers, who already have a considerable amount of data, clustered to a much lesser extent or not clustered at all. The main idea of this method is that applying clustering technique when the number of transactional data of the target customer is less than the predefined criterion data size. In order to find this criterion number, we suggest the algorithm called sliding window correlation analysis in this study. The algorithm purposes to find the transactional data size that the performance of the 1-to-1 method is radically decreased due to the data sparity. After finding this criterion data size, we apply the conventional 1-to-1 method for the customers who have more data than the criterion and apply clustering technique who have less than this amount until they can use at least the predefined criterion amount of data for model building processes. We apply the two conventional methods and the newly suggested method to Neilsen's beverage purchasing data to predict the purchasing amounts of the customers and the purchasing categories. We use two data mining techniques (Support Vector Machine and Linear Regression) and two types of performance measures (MAE and RMSE) in order to predict two dependent variables as aforementioned. The results show that the suggested Intelligent Customer Segmentation method can outperform the conventional 1-to-1 method in many cases and produces the same level of performances compare with the Customer-Segmentation method spending much less computational cost.

• 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.