Graphical causal inference and copula regression model for apple keywords by text mining

Jong Min Kim, Sunghae Jun

    Research output: Contribution to journalArticlepeer-review

    24 Scopus citations

    Abstract

    Apple is a leading company of technological evolution and innovation. This company founded and produced the Apple I computer in 1976. Since then, based on its innovative technologies, Apple has launched creative and innovative products and services such as the iPod, iTunes, the iPhone, the Apple app store, and the iPad. In many fields of academia and business, diverse studies of Apple's technological innovation strategy have been performed. In this paper, we analyze Apple's patents to better understand its technological innovation. We collected all applied patents by Apple until now, and applied statistics and text mining for patent analysis. By using graphical causal inference method, we created the causal relations among Apple keywords preprocessed by text mining, and then we carried out the semiparametric Gaussian copula regression model to see how the target response keyword and the predictor keywords are relating to each other. Furthermore, Gaussian copula partial correlation was applied to Apple keywords to find out the detailed dependence structure. By performing these methods, this paper shows the technological trends and relations between Apple's technologies. This research could make contributions in finding vacant technology areas and central technologies for Apple's R&D planning.

    Original languageEnglish (US)
    Pages (from-to)918-929
    Number of pages12
    JournalAdvanced Engineering Informatics
    Volume29
    Issue number4
    DOIs
    StatePublished - Oct 2015

    Keywords

    • Apple keywords
    • Copula regression
    • Graphical causal inference
    • Patent analysis
    • Text mining

    Fingerprint

    Dive into the research topics of 'Graphical causal inference and copula regression model for apple keywords by text mining'. Together they form a unique fingerprint.

    Cite this