• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Asia pacific journal of information systems
• /
• v.19 no.4
• /
• pp.125-148
• /
• 2009

UGC(User Generated Contents) is emerging as the center of e-business in the web 2.0 era. The trend reflects changing roles of users in production and consumption of contents on websites and helps us to understand new strategies of websites such as web portals and social network websites. Nowadays, we consume contents created by other non-professional users for both utilitarian (e.g., knowledge) and hedonic values (e.g., fun). Also, contents produced by ourselves (e.g., photo, video) are posted on websites so that our friends, family, and even the public can consume those contents. This means that non-professionals, who used to be passive audience in the past, are now creating contents and share their UGCs with others in the Web. Accessible media, tools, and applications have also reduced difficulty and complexity in the process of creating contents. Realizing that users create plenty of materials which are very interesting to other people, media companies (i.e., web portals and social networking websites) are adjusting their strategies and business models accordingly. Increased demand of UGC may lead to website visits which are the source of benefits from advertising. Therefore, they put more efforts into making their websites open platforms where UGCs can be created and shared among users without technical and methodological difficulties. Many websites have increasingly adopted new technologies such as RSS and openAPI. Some have even changed the structure of web pages so that UGC can be seen several times to more visitors. This mainstream of UGCs on websites indicates that acquiring more UGCs and supporting participating users have become important things to media companies. Although those companies need to understand why general users have shown increasing interest in creating and posting contents and what is important to them in the process of productions, few research results exist in this area to address these issues. Also, behavioral process in creating video UGCs has not been explored enough for the public to fully understand it. With a solid theoretical background (i.e., theory of implementation intentions), parts of our proposed research model mirror the process of user behaviors in creating video contents, which consist of intention to upload, intention to edit, edit, and upload. In addition, in order to explain how those behavioral intentions are developed, we investigated influences of antecedents from three motivational perspectives (i.e., intrinsic, editing software-oriented, and website's network effect-oriented). First, from the intrinsic motivation perspective, we studied the roles of self-expression, enjoyment, and social attention in forming intention to edit with preferred editing software or in forming intention to upload video contents to preferred websites. Second, we explored the roles of editing software for non-professionals to edit video contents, in terms of how it makes production process easier and how it is useful in the process. Finally, from the website characteristic-oriented perspective, we investigated the role of a website's network externality as an antecedent of users' intention to upload to preferred websites. The rationale is that posting UGCs on websites are basically social-oriented behaviors; thus, users prefer a website with the high level of network externality for contents uploading. This study adopted a longitudinal research design; we emailed recipients twice with different questionnaires. Guided by invitation email including a link to web survey page, respondents answered most of questions except edit and upload at the first survey. They were asked to provide information about UGC editing software they mainly used and preferred website to upload edited contents, and then asked to answer related questions. For example, before answering questions regarding network externality, they individually had to declare the name of the website to which they would be willing to upload. At the end of the first survey, we asked if they agreed to participate in the corresponding survey in a month. During twenty days, 333 complete responses were gathered in the first survey. One month later, we emailed those recipients to ask for participation in the second survey. 185 of the 333 recipients (about 56 percentages) answered in the second survey. Personalized questionnaires were provided for them to remind the names of editing software and website that they reported in the first survey. They answered the degree of editing with the software and the degree of uploading video contents to the website for the past one month. To all recipients of the two surveys, exchange tickets for books (about 5,000~10,000 Korean Won) were provided according to the frequency of participations. PLS analysis shows that user behaviors in creating video contents are well explained by the theory of implementation intentions. In fact, intention to upload significantly influences intention to edit in the process of accomplishing the goal behavior, upload. These relationships show the behavioral process that has been unclear in users' creating video contents for uploading and also highlight important roles of editing in the process. Regarding the intrinsic motivations, the results illustrated that users are likely to edit their own video contents in order to express their own intrinsic traits such as thoughts and feelings. Also, their intention to upload contents in preferred website is formed because they want to attract much attention from others through contents reflecting themselves. This result well corresponds to the roles of the website characteristic, namely, network externality. Based on the PLS results, the network effect of a website has significant influence on users' intention to upload to the preferred website. This indicates that users with social attention motivations are likely to upload their video UGCs to a website whose network size is big enough to realize their motivations easily. Finally, regarding editing software characteristic-oriented motivations, making exclusively-provided editing software more user-friendly (i.e., easy of use, usefulness) plays an important role in leading to users' intention to edit. Our research contributes to both academic scholars and professionals. For researchers, our results show that the theory of implementation intentions is well applied to the video UGC context and very useful to explain the relationship between implementation intentions and goal behaviors. With the theory, this study theoretically and empirically confirmed that editing is a different and important behavior from uploading behavior, and we tested the behavioral process of ordinary users in creating video UGCs, focusing on significant motivational factors in each step. In addition, parts of our research model are also rooted in the solid theoretical background such as the technology acceptance model and the theory of network externality to explain the effects of UGC-related motivations. For practitioners, our results suggest that media companies need to restructure their websites so that users' needs for social interaction through UGC (e.g., self-expression, social attention) are well met. Also, we emphasize strategic importance of the network size of websites in leading non-professionals to upload video contents to the websites. Those websites need to find a way to utilize the network effects for acquiring more UGCs. Finally, we suggest that some ways to improve editing software be considered as a way to increase edit behavior which is a very important process leading to UGC uploading.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.5 no.3
• /
• pp.200-206
• /
• 1985

This field experiment was undertaken to assess the effects of three levels of inclination ($10^{\circ},\;20^{\circ},\;and\;30^{\circ}$) and four rates of $N-P_2O_5-K_2O$ (0-0-0-, 14-10-10, 28-25-25, and 42-40-40kg/10a) on establishment, yield and quality, and botanical compositions of mixed grass-clover sward. This second part is concerned with the soil chemical properties, concentrations and uptake of mineral nutrients, and percent recovery and efficiency of NPK. The results obtained after a two-year experiment are summarized as follows: 1. The pH, exchangeable Mg and Na, and base saturation in the surface soils were decreased by increasing the grade of inclination, whereas organic matter and available $P_2O_5$ tended to be increased. However, the changes in the Ca content and equivalent ratio of $K\sqrt{Ca+Mg}$ were not significant. The pH, exchangeable Ca and Mg, and base saturation were reduced by increasing the NPK rate, whereas available $P_2O_5$, exchangeable K, and equivalent ratio of $K\sqrt{Ca+Mg}$ tended to be increased. 2. The concentrations of mineral nutrients in grasses and weeds were not significantly affected by increasing the grade of slope in hilly pasture, whereas the concentrations of N, K, and Mg in legume were the lowest with the steep slope, which seemed to be related to the low legume yield. The Mg concentrations of all forage species were below the critical level for good forage growth and likelihood of grass tetany. 3. The increase of NPK rate resulted in the increment of N, K and Na concentrations, and the decrease of Mg and Ca in grasses. The P concentration was increased with P application, but there were no differences in that among the P rates applied. It resulted also in a slight increase of K, and a decrease of Mg in legume, but the contents of N, Ca, and Na were not affected by that. On the other hand, it has not affected the mineral contents in weeds except a somewhat increase of N. The mixed forages showed a increase of N and K contents, a decrease of Ca and Mg, and a slight change in P and Na. 4. The percent recovery of N, P and K by mixed forages were greatly decreased by increasing the grade of inclination and NPK rate. They were high in the order; K>N>P. The efficiency of mixed NPK applications was decreased by that. The efficiency of mixed NPK fertilizers absorbed was slightly decreased by the increased rate of NPK, but it was not affected by the grade of inclination.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.31 no.2
• /
• pp.142-152
• /
• 1995

The combined bottom trawl and hydroacoustic survey was conducted by using the training ship Oshoro Maru belong to Hokkaido University in November 1989-1992 and the training ship Nagasaki Maru belong to Nagasaki University in April 1994 in the East China Sea, respectively. The aim of the investigations was to collect the target strength data of fish school in relation to the biomass estimation of fish in the survey area. The hydroacoustic survey was performed by using the scientific echo sounder system operating at three frequencies of 25, 50 and 100kHz with a microcomputer-based echo integrator. Fish samples were collected by bottom trawling and during the trawl surveys, the openings of otter board and net mouth were measured. The target strength of fish school was estimated from the relationship between the volume back scattering strength for the depth strata of bottom trawling and the weight per unit volume of trawl catches. A portion of the trawl catches preserved in frozon condition on board, the target strength measurements for the defrosted samples of ten species were conducted in the laboratory tank, and the relationship between target strength and fish weight was examined. In order to investigate the effect of swimbladder on target strength, the volume of the swimbladder of white croaker, Argyrosomus argentatus, sampled by bottom trawling was measured by directly removing the gas in the swimbladder with a syringe on board. The results obtained can be summarized as follows: 1.The relationship between the mean volume back scattering strength (, dB) for the depth strata of trawl hauls and the weight(C, $kg/\textrm{m}^3$) per unit volume of trawl catches were expressed by the following equations : 25kHz : = - 29.8+10Log(C) 50kHz : = - 32.4+10Log(C) 100kHz : = - 31.7+10Log(C) The mean target strength estimates for three frequencies of 25, 50 and 100 kHz derived from these equations were -29.8dB/kg, -32.4dB/kg and -31.7dB/kg, respectively. 2. The relationship between target strength and body weight for the fish samples of ten species collected by trawl surveys were expressed by the following equations : 25kHz : TS = - 34.0+10Log($W^{\frac{2}{3}}$) 100kHz : TS = - 37.8+10Log($W^{\frac{2}{3}}$) The mean target strength estimates for two frequencies of 25 and 100 kHz derived from these equations were -34.0dB/kg, -37.8dB/kg, respectively. 3. The representative target strength values for demersal fish populations of the East China Sea at two frequencies of 25 and 100 kHz were estimated to be -31.4dB/kg, -33.8dB/kg, respectively. 4. The ratio of the equivalent radius of swimbladder to body length of white croaker was 0.089 and the volume of swimbladder was estimated to be approximately 10% of total body volume.

• The Journal of Korean Society for Radiation Therapy
• /
• v.30 no.1_2
• /
• pp.191-199
• /
• 2018

Purpose : In this study, we evaluated the usefulness of colloid gel(slime) as a compensator for irregular patient surfaces in radiation therapy. Materials and Methods : For this study, colloid gel suitable for treatment was made and four experiments were conducted to evaluate the applicability of radiation therapy. Trilogy(Varian) and CT(SOMATOM, Siemens) were used as treatment equipment and CT equipment. First, the homogeneity according to the composition of colloid gel was measured using EBT3 Film(RIT). Second, the Hounsfield Unit(HU) value of colloid gel was measured and confirmed by CRIS phantom, Eclipse RTP(Eclipse 13.1, Varian) and CT. Third, to measure the deformation and degeneration of colloid gel during the treatment period, it was measured 3 times daily for 2 weeks using an ion chamber(PTW-30013, PTW). The fourth experiment was compared the treatment plan and measured dose distributions using bolus, rice, colloid gel and additional, dose profiles in an environment similar to actual treatment using our own acrylic phantom. Result : First experiment, density of the colloid gel cases 1, 2 and 3 was $1.02g/cm^3$, $0.99g/cm^3$ and $0.96g/cm^3$. When the homogeneity was measured at 6 MV and 9 MeV, case 1 was more homogeneous than the other cases, as 1.55 and 1.98. In the second experiment, the HU values of case 1, 2, 3 were 15 and when the treatment plan was compared with the measured doses, the difference was within 1 % at all 9, 12 MeV and a difference of -1.53 % and -1.56 % within the whole 2 % at 6 MV. In the third experiment, the dose change of colloid gel was measured to be about 1 % for 2 weeks. In the fourth experiment, the dose difference between the treatment plan and EBT3 film was similar for both colloid gel and bolus, rice at 6 MV. But colloid gel showed less dose difference than bolus and rice at 9 MeV. Also, dose profile of colloid gel showed a more uniform dose distribution than the bolus and rice. Conclusion : In this study, the density of colloid gel prepared for radiation therapy was $1.02g/cm^3$ similar to the density of water, and alteration or deformation was not observed during the radiotherapy process. Although we pay attention to the density when manufacturing colloid gel, it is sufficient in that it can deliver the dose uniformly through the compensation of the patient's body surface more than the bolus and rice, and can be manufactured at low cost. Further studies and studies for clinical applications are expected to be applicable to radiation therapy.

• The Journal of Korean Society for Radiation Therapy
• /
• v.30 no.1_2
• /
• pp.49-63
• /
• 2018

Purpose : The purpose of this study is investigating the changes of treatment plan and comparing skin dose with or without the skin flash. To investigate optimal applications of the skin flash, the changes of skin dose of each plans by various thicknesses of skin flash were measured and analyzed also. Methods and Material : Anthropomorphic phantom was scanned by CT for this study. The 2 fields hybrid IMRT and the 6 fields static IMRT were generated from the Eclipse (ver. 13.7.16, Varian, USA) RTP system. Additional plans were generated from each IMRT plans by changing skin flash thickness to 0.5 cm, 1.0 cm, 1.5 cm, 2.0 cm and 2.5 cm. MU and maximum doses were measured also. The treatment equipment was 6MV of VitalBeam (Varian Medical System, USA). Measuring device was a metal oxide semiconductor field-effect transistor(MOSFET). Measuring points of skin doses are upper (1), middle (2) and lower (3) positions from center of the left breast of the phantom. Other points of skin doses, artificially moved to medial and lateral sides by 0.5 cm, were also measured. Results : The reference value of 2F-hIMRT was 206.7 cGy at 1, 186.7 cGy at 2, and 222 cGy at 3, and reference values of 6F-sIMRT were measured at 192 cGy at 1, 213 cGy at 2, and 215 cGy at 3. In comparison with these reference values, the first measurement point in 2F-hIMRT was 261.3 cGy with a skin flash 2.0 cm and 2.5 cm, and the highest dose difference was 26.1 %diff. and 5.6 %diff, respectively. The third measurement point was 245.3 cGy and 10.5 %diff at the skin flash 2.5 cm. In the 6F-sIMRT, the highest dose difference was observed at 216.3 cGy and 12.7 %diff. when applying the skin flash 2.0 cm for the first measurement point and the dose difference was the largest at the application point of 2.0 cm, not the skin flash 2.5 cm for each measurement point. In cases of medial 0.5 cm shift points of 2F-hIMRT and 6F-sIMRT without skin flash, the measured value was -75.2 %diff. and -70.1 %diff. at 2F, At -14.8, -12.5, and -21.0 %diff. at the 1st, 2nd and 3rd measurement points, respectively. Generally, both treatment plans showed an increase in total MU, maximum dose and %diff as skin flash thickness increased, except for some results. The difference of skin dose using 0.5 cm thickness of skin flash was lowest lesser than 20 % in every conditions. Conclusion : Minimizing the thickness of skin flash by 0.5 cm is considered most ideal because it makes it possible to keep down MUs and lowering maximum doses. In addition, It was found that MUs, maximum doses and differences of skin doses did not increase infinitely as skin flash thickness increase by. If the error margin caused by PTV or other factors is lesser than 1.0 cm, It is considered that there will be many advantages in with the skin flash technique comparing without it.

• Microbiology and Biotechnology Letters
• /
• v.46 no.4
• /
• pp.360-371
• /
• 2018

Extracts and fractions of Anemarrhena asphodeloides Bunge were prepared and their physiological activities and components were analyzed. Antimicrobial activities of the ethyl acetate and aglycone fractions were $78{\mu}g/ml$ and $31{\mu}g/ml$, respectively, for Staphylococcus aureus and $156{\mu}g/ml$ and $125{\mu}g/ml$, respectively, for Pseudomonas aeruginosa. 1,1-Diphenyl-2-picrylhydrazyl free radical scavenging activities ($FSC_{50}$) of 50% ethanol extract, ethyl acetate fraction, and aglycone fraction of A. asphodeloides extracts were $146.2{\mu}g/ml$, $23.19{\mu}g/ml$, and $71.06{\mu}g/ml$, respectively. The total antioxidant capacity ($OSC_{50}$) in an $Fe^{3+}$-EDTA/hydrogen peroxide ($H_2O_2$) system were $17.5{\mu}g/ml$, $1.5{\mu}g/ml$, and $1.4{\mu}g/ml$, respectively. The cytoprotective effect (${\tau}_{50}$) in $^1O_2$-induced erythrocyte hemolysis was 181 min with $4{\mu}g/ml$ of the aglycone fraction. The ${\tau}_{50}$ of the aglycone fraction was approximately 4-times higher than that of (+)-${\alpha}$-tocopherol (${\tau}_{50}$, 41 min). Analysis of $H_2O_2$-induced damage of HaCaT cells revealed that the maximum cell viabilities for the 50% ethanol extract, ethyl acetate fraction, and aglycone fraction were 86.23%, 86.59%, and 89.70%, respectively. The aglycone fraction increased cell viability up to 11.53% at $1{\mu}g/ml$ compared to the positive control treated with $H_2O_2$. Analysis of ultraviolet B radiation-induced HaCaT cell damage revealed up to 41.77% decreased intracellular reactive oxygen species in the $2{\mu}g/ml$ aglycone fraction compared with the positive control treated with ultraviolet B radiation. The findings suggest that the extracts and fractions of A. asphodeloides Bunge have potential applications in the field of cosmetics as natural preservatives and antioxidants.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.1-32
• /
• 2018

In addition to stakeholders including managers, employees, creditors, and investors of bankrupt companies, corporate defaults have a ripple effect on the local and national economy. Before the Asian financial crisis, the Korean government only analyzed SMEs and tried to improve the forecasting power of a default prediction model, rather than developing various corporate default models. As a result, even large corporations called 'chaebol enterprises' become bankrupt. Even after that, the analysis of past corporate defaults has been focused on specific variables, and when the government restructured immediately after the global financial crisis, they only focused on certain main variables such as 'debt ratio'. A multifaceted study of corporate default prediction models is essential to ensure diverse interests, to avoid situations like the 'Lehman Brothers Case' of the global financial crisis, to avoid total collapse in a single moment. The key variables used in corporate defaults vary over time. This is confirmed by Beaver (1967, 1968) and Altman's (1968) analysis that Deakins'(1972) study shows that the major factors affecting corporate failure have changed. In Grice's (2001) study, the importance of predictive variables was also found through Zmijewski's (1984) and Ohlson's (1980) models. However, the studies that have been carried out in the past use static models. Most of them do not consider the changes that occur in the course of time. Therefore, in order to construct consistent prediction models, it is necessary to compensate the time-dependent bias by means of a time series analysis algorithm reflecting dynamic change. Based on the global financial crisis, which has had a significant impact on Korea, this study is conducted using 10 years of annual corporate data from 2000 to 2009. Data are divided into training data, validation data, and test data respectively, and are divided into 7, 2, and 1 years respectively. In order to construct a consistent bankruptcy model in the flow of time change, we first train a time series deep learning algorithm model using the data before the financial crisis (2000~2006). The parameter tuning of the existing model and the deep learning time series algorithm is conducted with validation data including the financial crisis period (2007~2008). As a result, we construct a model that shows similar pattern to the results of the learning data and shows excellent prediction power. After that, each bankruptcy prediction model is restructured by integrating the learning data and validation data again (2000 ~ 2008), applying the optimal parameters as in the previous validation. Finally, each corporate default prediction model is evaluated and compared using test data (2009) based on the trained models over nine years. Then, the usefulness of the corporate default prediction model based on the deep learning time series algorithm is proved. In addition, by adding the Lasso regression analysis to the existing methods (multiple discriminant analysis, logit model) which select the variables, it is proved that the deep learning time series algorithm model based on the three bundles of variables is useful for robust corporate default prediction. The definition of bankruptcy used is the same as that of Lee (2015). Independent variables include financial information such as financial ratios used in previous studies. Multivariate discriminant analysis, logit model, and Lasso regression model are used to select the optimal variable group. The influence of the Multivariate discriminant analysis model proposed by Altman (1968), the Logit model proposed by Ohlson (1980), the non-time series machine learning algorithms, and the deep learning time series algorithms are compared. In the case of corporate data, there are limitations of 'nonlinear variables', 'multi-collinearity' of variables, and 'lack of data'. While the logit model is nonlinear, the Lasso regression model solves the multi-collinearity problem, and the deep learning time series algorithm using the variable data generation method complements the lack of data. Big Data Technology, a leading technology in the future, is moving from simple human analysis, to automated AI analysis, and finally towards future intertwined AI applications. Although the study of the corporate default prediction model using the time series algorithm is still in its early stages, deep learning algorithm is much faster than regression analysis at corporate default prediction modeling. Also, it is more effective on prediction power. Through the Fourth Industrial Revolution, the current government and other overseas governments are working hard to integrate the system in everyday life of their nation and society. Yet the field of deep learning time series research for the financial industry is still insufficient. This is an initial study on deep learning time series algorithm analysis of corporate defaults. Therefore it is hoped that it will be used as a comparative analysis data for non-specialists who start a study combining financial data and deep learning time series algorithm.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.21-41
• /
• 2019

Thanks to the rapid development of information technologies, the data available on Internet have grown rapidly. In this era of big data, many studies have attempted to offer insights and express the effects of data analysis. In the tourism and hospitality industry, many firms and studies in the era of big data have paid attention to online reviews on social media because of their large influence over customers. As tourism is an information-intensive industry, the effect of these information networks on social media platforms is more remarkable compared to any other types of media. However, there are some limitations to the improvements in service quality that can be made based on opinions on social media platforms. Users on social media platforms represent their opinions as text, images, and so on. Raw data sets from these reviews are unstructured. Moreover, these data sets are too big to extract new information and hidden knowledge by human competences. To use them for business intelligence and analytics applications, proper big data techniques like Natural Language Processing and data mining techniques are needed. This study suggests an analytical approach to directly yield insights from these reviews to improve the service quality of hotels. Our proposed approach consists of topic mining to extract topics contained in the reviews and the decision tree modeling to explain the relationship between topics and ratings. Topic mining refers to a method for finding a group of words from a collection of documents that represents a document. Among several topic mining methods, we adopted the Latent Dirichlet Allocation algorithm, which is considered as the most universal algorithm. However, LDA is not enough to find insights that can improve service quality because it cannot find the relationship between topics and ratings. To overcome this limitation, we also use the Classification and Regression Tree method, which is a kind of decision tree technique. Through the CART method, we can find what topics are related to positive or negative ratings of a hotel and visualize the results. Therefore, this study aims to investigate the representation of an analytical approach for the improvement of hotel service quality from unstructured review data sets. Through experiments for four hotels in Hong Kong, we can find the strengths and weaknesses of services for each hotel and suggest improvements to aid in customer satisfaction. Especially from positive reviews, we find what these hotels should maintain for service quality. For example, compared with the other hotels, a hotel has a good location and room condition which are extracted from positive reviews for it. In contrast, we also find what they should modify in their services from negative reviews. For example, a hotel should improve room condition related to soundproof. These results mean that our approach is useful in finding some insights for the service quality of hotels. That is, from the enormous size of review data, our approach can provide practical suggestions for hotel managers to improve their service quality. In the past, studies for improving service quality relied on surveys or interviews of customers. However, these methods are often costly and time consuming and the results may be biased by biased sampling or untrustworthy answers. The proposed approach directly obtains honest feedback from customers' online reviews and draws some insights through a type of big data analysis. So it will be a more useful tool to overcome the limitations of surveys or interviews. Moreover, our approach easily obtains the service quality information of other hotels or services in the tourism industry because it needs only open online reviews and ratings as input data. Furthermore, the performance of our approach will be better if other structured and unstructured data sources are added.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.85-107
• /
• 2019

Online consumers browse products belonging to a particular product line or brand for purchase, or simply leave a wide range of navigation without making purchase. The research on the behavior and purchase of online consumers has been steadily progressed, and related services and applications based on behavior data of consumers have been developed in practice. In recent years, customization strategies and recommendation systems of consumers have been utilized due to the development of big data technology, and attempts are being made to optimize users' shopping experience. However, even in such an attempt, it is very unlikely that online consumers will actually be able to visit the website and switch to the purchase stage. This is because online consumers do not just visit the website to purchase products but use and browse the websites differently according to their shopping motives and purposes. Therefore, it is important to analyze various types of visits as well as visits to purchase, which is important for understanding the behaviors of online consumers. In this study, we explored the clustering analysis of session based on click stream data of e-commerce company in order to explain diversity and complexity of search behavior of online consumers and typified search behavior. For the analysis, we converted data points of more than 8 million pages units into visit units' sessions, resulting in a total of over 500,000 website visit sessions. For each visit session, 12 characteristics such as page view, duration, search diversity, and page type concentration were extracted for clustering analysis. Considering the size of the data set, we performed the analysis using the Mini-Batch K-means algorithm, which has advantages in terms of learning speed and efficiency while maintaining the clustering performance similar to that of the clustering algorithm K-means. The most optimized number of clusters was derived from four, and the differences in session unit characteristics and purchasing rates were identified for each cluster. The online consumer visits the website several times and learns about the product and decides the purchase. In order to analyze the purchasing process over several visits of the online consumer, we constructed the visiting sequence data of the consumer based on the navigation patterns in the web site derived clustering analysis. The visit sequence data includes a series of visiting sequences until one purchase is made, and the items constituting one sequence become cluster labels derived from the foregoing. We have separately established a sequence data for consumers who have made purchases and data on visits for consumers who have only explored products without making purchases during the same period of time. And then sequential pattern mining was applied to extract frequent patterns from each sequence data. The minimum support is set to 10%, and frequent patterns consist of a sequence of cluster labels. While there are common derived patterns in both sequence data, there are also frequent patterns derived only from one side of sequence data. We found that the consumers who made purchases through the comparative analysis of the extracted frequent patterns showed the visiting pattern to decide to purchase the product repeatedly while searching for the specific product. The implication of this study is that we analyze the search type of online consumers by using large - scale click stream data and analyze the patterns of them to explain the behavior of purchasing process with data-driven point. Most studies that typology of online consumers have focused on the characteristics of the type and what factors are key in distinguishing that type. In this study, we carried out an analysis to type the behavior of online consumers, and further analyzed what order the types could be organized into one another and become a series of search patterns. In addition, online retailers will be able to try to improve their purchasing conversion through marketing strategies and recommendations for various types of visit and will be able to evaluate the effect of the strategy through changes in consumers' visit patterns.