Sangsung Park

Expert system based on artificial neural networks for content-based image retrieval

Clustering technique is essential for fast retrieval in large database. In this paper, new image clustering technique based on artificial neural networks is proposed for content-based image retrieval. Fuzzy-ART mechanism maps high-dimensional input features into the output neuron. Joint HSV histogram and average entropy computed from gray-level co-occurrence matrices in the localized image region is employed as input feature elements. Original Fuzzy-ART suffers unnecessary increase of the number of output neurons when the noise input is presented. Modified Fuzzy-ART mechanism resolves the problem by differently updating the committed node and uncommitted node, and checking the vigilance test again. To show the validity of the proposed algorithm, experiment results on image clustering performance and comparison with original Fuzzy-ART are presented in terms of recall rates.

Methodology of technological evolution for three-dimensional printing

Purpose – An increasing amount of attention is being paid to three-dimensional (3D) printing technology. The technology itself is based on diverse technologies such as laser beams and materials. Hence, 3D printing technology is a converging technology that produces 3D objects using a 3D printer. To become technologically competitive, many companies and nations are developing technologies for 3D printing. So to know its technological evolution is meaningful for developing 3D printing in the future. The paper aims to discuss these issues. Design/methodology/approach – To get technological competitiveness of 3D printing, the authors should know the most important and essential technology for 3D printing. An understanding of the technological evolution of 3D printing is needed to forecast its future technologies and build the R & D planning needed for 3D printing. In this paper, the authors propose a methodology to analyze the technological evolution of 3D printing. The authors analyze entire patent documents...

Forecasting Vacant Technology of Patent Analysis System using Self Organizing Map and Matrix Analysis

Patent analysis is the extracting knowledge which is needed for the company`s research and development strategy through accumulated worldwide patent database. In order to set the future direction of corresponding technology which is scheduled to be developed, the technology trends and deployment processes are identified by analyzing results of present patent applications. The patent analysis provides the required results for analyzing present patent applications. In this paper, we will carry out technology classification for related patent analysis methods and systems. Moreover we will investigate and analyze related domestic patents, U.S. patents and IEEE papers. Due to the characteristics of technology sector, not only patents are applied but also research papers are released actively about patent analysis system. We will analyze patents according to the technology classification by using the final searching results which come from the selected search words in this study. To find necessary niche technology which is needed for patent analysis system, matrix analysis was performed to all of valid patents and papers. Identifying the technology development trends of registered patent analysis systems, and presenting the future direction of technology development which is related to patent analysis system. To figure out the technology which is developed relatively weak based on domestic patents, U.S patent and research papers by analyzing the valid patents and papers with statistical test and self-organizing map quantitatively. Then, presenting the necessity of this technology development.

Fuzzy Art-Based Image Clustering Method For Content-Based Image Retrieval

In this paper, an image clustering method that is essential for content-based image retrieval in large image databases efficiently is proposed by color, texture, and shape contents. The dominant triple HSV (Hue, Saturation, and Value), which are extracted from quantized HSV joint histogram in the image region, are used for representing color information in the image. Entropy and maximum entry from co-occurrence matrices are used for texture information and edge angle histogram is used for representing shape information. Due to its algorithmic simplicity and the several merits that facilitate the implementation of the neural network, Fuzzy ART has been exploited for image clustering. Original Fuzzy ART suffers unnecessary increase of the number of output neurons when the noise input is presented. Therefore, the improved Fuzzy ART algorithm is proposed to resolve the problem by differently updating the committed node and uncommitted node, and checking the vigilance test again. To show the validity of the proposed algorithm, experimental results on image clustering performance and comparison with original Fuzzy ART are presented in terms of recall rates.

Examining technological competition between BMW and Hyundai in the Korean car market

ABSTRACT The Korean car market has increased in size. BMW and Hyundai are the top-ranked imported and domestic vehicle brands in Korea, respectively. Thus, it is important to understand these companies and the Korean car market, because technology is most significant in the vehicle industry. In this paper, we compare BMW with Hyundai from the technological perspective. Our research is focused on an analysis of the technological competition between BMW and Hyundai based on their developed technologies. We use all BMW and Hyundai patents from worldwide patent databases to analyse the two companies’ technologies. In addition, we apply statistical methods and machine learning algorithms to the patent analysis. In our conclusion, we show the technological differences and competition between BMW and Hyundai, and find their relative strengths and weaknesses.

A Technology Valuation Model Using Quantitative Patent Analysis: A Case Study of Technology Transfer in Big Data Marketing

Abstract Technology valuation (TV) is an important issue in management of technology (MOT). We use TV results for technology transfer, research and development (R&D) planning, and technology marketing. Diverse TV studies have been applied to MOT. Most of them were dependent on domain experts’ knowledge, so their TV results could be subjective and unstable. To solve this problem, we propose an objective TV model using quantitative patent analysis. In this article, we consider text mining, social network analysis, technology clustering, and descriptive statistics in constructing our TV model. To verify the performance of our model, we perform a case study of technology transfer in big data marketing.

A Study on Development of Patent Information Retrieval Using Textmining

The patent information retrieval system can serve a variety of purposes. In general, the patent information is retrieved using limited key words. To identify earlier technology and priority rights repeated effort is needed. This study proposes a method of content-based retrieval using text mining. Using the proposed algorithm, each of the documents is invested with characteristic value. The characteristic values are used to compare similarities between query documents and database documents. Text analysis is composed of 3 steps: stop-word, keyword analysis and weighted value calculation. In the test results, the general retrieval and the proposed algorithm were compared by using accuracy measurements. As the study arranges the result documents as similarities of the query documents, the surfer can improve the efficiency by reviewing the similar documents first. Also because of being able to input the full-text of patent documents, the users unacquainted with surfing can use it easily and quickly. It can reduce the amount of displayed missing data through the use of content based retrieval instead of keyword based retrieval for extending the scope of the search.

Establishment of Strategy for Management of Technology Using Data Mining Technique

Abstract Technology forecasting is about understanding a status of a specific technology in the future, based on the cur-rent data of the technology. It is useful when planning technology management strategies. These days, it is com-mon for countries, companies, and researchers to establish R&D directions and strategies by utilizing experts’ opinions. However, this qualitative method of technology foreca sting is costly and time consuming since it re-quires to collect a variety of opinions and analysis from many experts. In order to deal with these limitations, quantitative method of technology forecasting is being studied to secure objective forecast result and help R&D decision making process. This paper suggests a methodology of t echnology forecasting based on quantitative analysis. The methodology consists of data collection, principa l component analysis, and technology forecasting by logistic regression, which is one of the data mining techniq ues. In this research, patent documents related toautonomous vehicle are collected. Then, the texts from patent d ocuments are extracted by text mining technique to construct an appropriate form for analysis. After principal component analysis, logistic regression is performedby using principal component score. On the basis of this result , it is possible to analyze R&D development sit-uation and technology forecasting. Key Words : Technology Forecasting, Patent, Logistic Regression, Autonoum ous vehicle, Principal component analysisReceived: Sep. 14, 2014Revised : Sep. 28, 2014Accepted: Jan. 20, 2015

Patent Big Data Analysis using Fuzzy Learning

Big data has had an immense effect on most social and industrial fields. It has three main characteristics, namely volume, variety, and velocity. Volume refers to the tremendous size of big data, variety pertains to its heterogeneous sources including numbers, text, and figures, and velocity refers to the rapid speed of data growth. Patent documents follow the characteristics of big data. A patent contains various results about the developed technology such as title, abstract, citations, figures, and drawings. In general, the volume of patent documents related to a target technology is very large. Moreover, a massive number of patent applications are submitted to the patent offices in every country daily. Patent data are analyzed for R&D planning by many institutes and companies. In this study, we propose a methodology for technology analysis applied to patent big data. Additionally, we employ fuzzy learning based on the fuzzy rule-based system for patent big data analysis. We study the fuzzy models for classification, regression, and clustering and group the patents by the fuzzy classification model. Using a fuzzy regression model, we build a technological relationship between subtechnologies. Lastly, we develop a fuzzy clustering model for technology clustering. To illustrate how our research may be applied to a practical domain, we employ a case study using the patent documents related to the three-dimensional printing technology.

A Novel Methodology for Extracting Core Technology and Patents by IP Mining

Abstract Society has been developed through analogue, digital, and smart era. Every technology is going through consistent changes and rapid developments. In this competitive society, R&D strategy establishment is sig-nificantly useful and helpful for improving technology competitiveness. A patent document includes technical and legal rights information such as title, abstract, description, claim, and patent classification code. From the paetn dtocumen,t a ol otf peopel can undersatnd and coellc ltega land techncia lniformaoitn. Thsi unqiue feature of patent can be quantitatively applied for technology analysis. This research paper proposes a meth-odology for extracting core technology and patents based on quantitative methods. Statistical analysis and so-cai nlewtork anaylssi are appeild ot IPC codes ni order to exrtac tcore technoolgies whti acvite R&D and high centralities. Then, core patents are also extracted by analyzing citation and family information.Key Words : Patent analysis, IPC Code, Social Network Analysis, Patent Citation, Patent FamilyReceived: Mar. 22, 2015Revised : Apr. 5, 2015Accepted: Jun. 1, 2015