Starting from Patents to Find Inputs to the Problem Graph Model of IDM-TRIZ

Souili, Achille; Cavallucci, Denis; Rousselot, François; Zanni, Cecilia

doi:10.1016/j.proeng.2015.12.365

Cited by 32 publications

(18 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In (Wang, 2015), the SAO structures are extracted based on the rules using the Stanford parser software, but there are no ready-made models for the Russian language. Papers (Guo, 2016, Souili, 2015 process patents using linguistic markers (specific verbs and nouns) and lexical-syntactic patterns, while (Souili, 2015) notes the need to study the structure of patent documents to improve the quality of data extraction. The emphasis is mainly on rule-based systems, since for statistical analysis systems a lot of marked-up data is obviously needed.…”

Section: Research Backgroundmentioning

confidence: 99%

Extraction of Structural Elements of Inventions From Russian-Language Patents

Korobkin¹,

Vasiliev²,

Fomenkov³

et al. 2019

Proceedings of the International Conferences Big Data Analytics, Data Mining and Computational Intelligence 2019; And Theory An

View full text Add to dashboard Cite

The paper investigates the issue of extracting structured data from Russian-language patents in the field of new technical solution synthesis. The predicate-argument constructions that characterize the composition of the structural elements of the inventions and the relations between them are the extracted entities. The paper also analyses existing natural language processing tools as applied to patent processing. The authors have offered a new method for extracting predicate-argument constructs taking into account the specificity of the text of patents based on shallow parsing and segmentation of sentences. The value of the F1 metric with a rigorous and lax evaluation of data extraction is 63% and 79%, respectively, which suggests that the proposed approach is promising. The extracted structures are used to construct the graph of the structural elements of the invention.

show abstract

Section: Research Backgroundmentioning

confidence: 99%

Extraction of Structural Elements of Inventions From Russian-Language Patents

Korobkin¹,

Vasiliev²,

Fomenkov³

et al. 2019

Proceedings of the International Conferences Big Data Analytics, Data Mining and Computational Intelligence 2019; And Theory An

View full text Add to dashboard Cite

show abstract

“…Souili and Cavallucci (2012) extract, from patent documents, knowledge such as problems, partial solutions and parameters. Using linguistic markers and NLP techniques, the authors match and extract knowledge relevant to the IDM ontology (Souili, Cavallucci, and Rousselot 2015b), (Souili et al 2015).…”

Section: Using Knowledge: Overview Of Various Patent Exploitation Metmentioning

confidence: 99%

A new method for extracting knowledge from patents to inspire designers during the problem-solving phase

Valverde

Nadeau

Scaravetti

2017

Journal of Engineering Design

View full text Add to dashboard Cite

Nowadays, patents are more than an industrial property protection, they constitute a considerable source of information. Several approaches deal with the extraction of pertinent information from patents, some of them translate that information into useful knowledge for problem-solving purposes. However, very few methods use a physical approach and a consistent analysis of the main function in the problem definition phase. This paper proposes a new method to inspire designers with ideas and analogous solutions. The objective is to contribute to solve complex technical problems by exploiting knowledge found in patent documents. The method is divided into three sections: problem definition, research and analysis, and innovation. The first two sections use iteratively a knowledge base formed by pertinent keywords related to the problem context, physical keywords from functional decomposition and physical analysis, and technological keywords recovered through patent analysis. In the third section, interesting patents are classified in a structured discovery matrix from the physical phenomena involved which are crossed in accordance with the related techniques found. The method is supported by a physical effects database of energy conversion and by evolution trends of technical systems. A deep offshore biphasic separator illustrates the method.

show abstract

“…Using textual analysis of patents for use in TRIZ, Cascini and Russo presented a way to automatically identifying the contradiction underlying a given technical system [70]. To identify relevant candidates for TRIZ automatically, Souili et al developed a method using linguistic markers [71,72].…”

Section: Patent Datasetsmentioning

confidence: 99%

Function Based Design-by-Analogy: A Functional Vector Approach to Analogical Search

Murphy

Otto

et al. 2014

Journal of Mechanical Design

100

View full text Add to dashboard Cite

Design-by-analogy is a powerful approach to augment traditional concept generation methods by expanding the set of generated ideas using similarity relationships from solutions to analogous problems. While the concept of design-by-analogy has been known for some time, few actual methods and tools exist to assist designers in systematically seeking and identifying analogies from general data sources, databases, or repositories, such as patent databases. A new method for extracting functional analogies from data sources has been developed to provide this capability, here based on a functional basis rather than form or conflict descriptions. Building on past research, we utilize a functional vector space model (VSM) to quantify analogous similarity of an idea's functionality. We quantitatively evaluate the functional similarity between represented design problems and, in this case, patent descriptions of products. We also develop document parsing algorithms to reduce text descriptions of the data sources down to the key functions, for use in the functional similarity analysis and functional vector space modeling. To do this, we apply Zipf's law on word count order reduction to reduce the words within the documents down to the applicable functionally critical terms, thus providing a mapping process for function based search. The reduction of a document into functional analogous words enables the matching to novel ideas that are functionally similar, which can be customized various ways. This approach thereby provides relevant sources of design-byanalogy inspiration. As a verification of the approach, two original design problem case studies illustrate the distance range of analogical solutions that can be extracted. This range extends from very near-field, literal solutions to far-field cross-domain analogies.

show abstract

Starting from Patents to Find Inputs to the Problem Graph Model of IDM-TRIZ

Cited by 32 publications

References 10 publications

Extraction of Structural Elements of Inventions From Russian-Language Patents

Extraction of Structural Elements of Inventions From Russian-Language Patents

A new method for extracting knowledge from patents to inspire designers during the problem-solving phase

Function Based Design-by-Analogy: A Functional Vector Approach to Analogical Search

Contact Info

Product

Resources

About