[Pycon] [new paper] "Maddalena Agnoli" - Python, patents and standards - An algorithme to classify and relate formal knowledge

[info a pycon.it]

Title: Python, patents and standards - An algorithme to classify and relate formal knowledge
Duration: 45 (includes Q&A)
Q&A Session: 15
Language: en
Type: Talk

Abstract: Economic research has become increasingly dependent on large scale data analysis. Economists spend more and more time on learning and applying programming languages in order to deal with the large amount of available data.

Innovation economics covers all kinds of topics related to innovative activities in an economy and during the last twenty years, scholars have started to use patent documents in order to capture innovativeness. Patent documents are text documents which describe a patent technology in detail. With the use of these data, the need for textual analysis, web scraping, big data analysis, etc. arose. In my research, I used Python for several reasons. A patent document contains valuable information about a technology and tells us furthermore who invented, owns, or traded this technology.

I use textual analysis in Python in order to create technology classifications for patents and compare them among each other, but also with other documents such as standard documents. Technological standards describe technologies which have been adopted as a general rule in a sector (e.g. Wi-Fi, USB, etc.). Using textual analysis, graph theory, and statistical analysis, I developed a technological concordance between patents and standards. I aim to present the use of data mining and analytic tools in Python for economic research in general, as well as the algorithm I developed to match patents to technologies adopted in standards.

Tags: [u'data-mining', u'text-analysis', u'DataAnalysis', u'text_mining']