Yoo-Jin Moon

Syntactic and semantic disambiguation of numeral strings using an n-gram method

This paper describes the interpretation of numerals, and strings including numerals, composed of a number and words or symbols that indicate whether the string is a SPEED, LENGTH, or whatever. The interpretation is done at three levels: lexical, syntactic, and semantic. The system employs three interpretation processes: a word trigram constructor with tokeniser, a rule-based processor of number strings, and n-gram based disambiguation of meanings. We extracted numeral strings from 378 online newspaper articles, finding that, on average, they comprised about 2.2% of the words in the articles. We chose 287 of these articles to provide unseen test data (3251 numeral strings), and used the remaining 91 articles to provide 886 numeral strings for use in manually extracting n-gram constraints to disambiguate the meanings of the numeral strings. We implemented six different disambiguation methods based on category frequency statistics collected from the sample data and on the number of word trigram constraints of each category. Precision ratios for the six methods when applied to the test data ranged from 85.6% to 87.9%.

Concept based image retrieval using the domain ontology

The recent study has been progressed the research about more semantic image indexing and retrieval. In our paper, we represent the improved concept-based image retrieval by using domain ontology. We analyze the many studies that applied the theory of ontology to concept-based image retrieval. Then, we try to solve the problems when we apply the huge ontologies in image retrieval system. There are two big problems. First, the huge ontologies that have many concepts, especially in particular domain, cannot express in existing ontologies. Therefore, in this paper we try to design and implement the domain ontology about the car based on the WordNet, which is one kinds of ontologies. The experimental result shows that the semantic distances between words are quite close when we test domain ontology than the existing WordNet.

A Study of Effects of UTAUT-Based Factors on Acceptance of Smart Health Care Services

The paper analyzes factors influencing acceptance of smart health care services based on UTAUT. The result of T-test indicates that users with experiences of smart health care services have a higher degree of effort expectancy and intention to use the services than those without their experiences. And, the study shows that social influence of the services positively affects user intention to use the services, that performance expectancy is positively correlated with user intention to use the services, and that perceived enjoyment positively affects potential intention to use the services. According to the results, companies need to increase performance expectancy, to intensify word-of-mouth marketing, and to improve enjoyment and attractiveness of the services.

Preferred Document Classification for a Highly Inflectional/Derivational Language

This paper describes methods of document classification for a highly inflectional/derivational language that forms monolithic compound noun terms, like Dutch and Korean. The system is composed of three phases: (1) a Korean morphological analyzer called HAM (Kang, 1993), (2) an application of compound noun phrase analysis to the result of HAM analysis and extraction of terms whose syntactic categories are noun, name (proper noun), verb, and adjective, and (3) an effective document classification algorithm based on preferred class score heuristics. This paper focuses on the comparison of document classification methods including a simple heuristic method, and preferred class score heuristics employing two factors namely ICF (inverted class frequency) and IDF (inverted document frequency) with/without term frequency weight. In addition this paper describes a simple classification approach without a learning algorithm rather than a vector space model with a complex training and classification algorithm such as cosine similarity measurement. The experimental results show 95.7% correct classifications of 720 training data and 63.8%-71.3% of randomly chosen 80 testing data through various methods.

An Empirical Study of Impacts of User Intention for Smart Wearable Devices and Use Behavior

The paper utilized the method of questionnaires, and on the empirical level analyzed impacts on intention to use smart wearable devices and use behavior based on UTAUT. The analysis showed that intention to use smart wearable devices depended on the level of performance expected by the consumer in utilizing smart wearable devices, on the hedonic experiences that the consumers enjoy, on the social influence that the consumer referents exert, and on the facilitating conditions available. Also it indicated that the actual use of smart wearable devices depended on the intention to use and the facilitating conditions. For management and marketing strategies of smart wearable device providers, implications of the two factors of hedonic motivation and performance expectancy were that consumers should experience the devices with enjoyment and get benefits by utilizing them. And the marketing strategies should appeal to consumers by positioning the using experience as an adventure or a way to reduce their stress and change a negative mood.

Adding the Temporal Relations in Semantic Web Ontologies

In studying the semantic web, the main area is to build web ontologies and create the relationships between concepts. Until now, the capabilities of markup languages for expressing the relationships have improved a great deal but it is insufficient for representing the temporal relationships. In this paper, we define the new axioms for representing the temporal relationships.

Korean compound noun term analysis based on a chart parsing technique

Unlike compound noun terms in English and French, where words are separated by white space, Korean compound noun terms are not separated by white space. In addition, some compound noun terms in the real world result from a spacing error. Thus the analysis of compound noun terms is a difficult task in Korean NLP. Systems based on probabilistic and statistical information extracted from a corpus have shown good performance on Korean compound noun analysis. However, if the domain of the actual system is expanded beyond that of the training system, then the performance on the compound noun analysis would not be consistent. In this paper, we will describe the analysis of Korean compound noun terms based on a longest substring algorithm and an agenda-based chart parsing technique, with a simple heuristic method to resolve the analyses’ ambiguities. The system successfully analysed 95.6% of the testing data (6024 compound noun terms) which ranged from 2 to 11 syllables. The average ambiguities ranged from 1 to 33 for each compound noun term.

Statistical Analysis of Determinants of Intention to Use Virtual Reality Services and Moderating Effects

This research statistically analyzes factors that influence consumer intention to use virtual reality services. The result shows that the main predictors of intention to use virtual reality services, in the order of importance, are hedonic motivation, personal innovativeness, effort expectancy and performance expectancy. And it shows that the higher were the impacts of effort expectancy, social influence, performance expectancy, and hedonic motivation on intention to use the services the higher was a customer’s personal innovativeness. According to the results, marketing strategies for virtual reality services should appeal to consumers by positioning the using experience as an adventure or a way to reduce their stress and change a negative mood. Also they should be reputation-building and target early adopters.

Forecasting Cultivable Region-Specific Crops Based on Future Climate Change Utilizing Public Big Data

The study designs and implements a database system for predicting small region-specific cultivable crops based on future climate change utilizing integrated public Big Data. For this study, regional temperature factors, regional precipitation factors, land acidity, solar radiation, cloud amount, and appropriate climatic factors for each crop were utilized. The database system could extract the information of each small region such as kinds of currently cultivating crop, kinds of regional cultivable food crop, kinds of regional cultivable fruit, kinds of regional cultivable medicinal crop, kinds of regional cultivable vegetable, and changing trends of each crop production quantity. Based on these small region-specific crop information, it is possible for the farmers to increase future profits of farm households by providing information of medicinal crops, food crops, vegetables, and fruits that can be produced in each regional farmhouse. It is also possible to present future recommended crops to individual business operators by utilizing these public big data, to suggest the need for development and research on crops that can be cultivated in each region, and to suggest marketing plans for present and future crops.

Impacts of Linked URLs in Social Media

This paper describes preliminary analysis of health-related social media postings in Twitter. We classified Tweets two ways: those (A) with and (B) without linked URLs, and similarly for users, those commonly posting Category A Tweets and users commonly posting Category B Tweets. The Tweet user groups and the two categories of Tweets show different characteristics in use of user-defined hash-tag terms, the impact of expanded URLs through social media community, and posting period of the two Tweet categories. One user among the top 25 most frequent posters in each user group only posted both Category A and B Tweets. Seven hash-tag terms from the top 25 hash-tag terms obtained from each category were used for both Category A and B. Tweets with and without linked URLs show different characteristics in terms of user groups, hash-tag terms, and the posting period of the linked URLs.