Mining layered technological information in scientific papers: A semi-supervised method

Abstract

Tech mining is the application of text mining tools to science and technology information resources. The ever-increasing volume of scientific outputs is a boom to technological innovation, but it also complicates efforts to obtain useful and concise information for problem solving. This challenge extends to tech mining, where the development of techniques compatible with big data is an urgent issue. This article introduces a semi-supervised method for extracting layered technological information from scientific papers in order to extend the reach of tech mining. Our method starts with several pre-set seed patterns used to extract candidate phrases by matching the dependency tree of each sentence. Then, after a series of judgements, phrases are divided into two categories: ‘main technique’ and ‘tech-component’. (A technique, for the purposes of this study, is a method or tool used in the article being analysed.) In order to generate new patterns for subsequent iterations, a weighted pattern learning method is also adopted. Finally, multiple iterations of the method are applied to extract technological information from each paper. A dataset from the field of optical switcher is used to verify the method’s effectiveness. Our findings are that (1) by two loops of extraction process in each iteration, our method realises the layered technological information extraction, which contains the ‘part–whole’ relationships between main techniques and tech-components; (2) the recall rate for main techniques is superior to the baseline after iterating 23 rounds; (3) when layering is disregarded, in the aspect of the precision and the volume of techniques, the new method is higher than that for the baseline; and (4) adjusting another two parameters can optimise the efficiency – however, the effect is neither pronounced nor straightforward.

Semantic TRIZ feasibility in technology development, innovation, and production: A systematic review

“… Technical feature ontology WIPO binary-syntactic relationships List of lemmas pool of Syntactic Dependency Pattern Unsupervised coarse classification Coarse Classification Precision Recall A37. A semi-supervised approach for analysing layered technical knowledge in scientific publications [ 65 ]. problem solving Extend to tech mining WoS database Tokenization (SAO) POS tagging Cohen's Kappa Coefficient (Statistic) Precision Recall F-score A38.…”

“… Ontological Approaches and System Modelling Exploration of ontological approaches, fact-oriented modelling, and system analysis integrating TRIZ principles with AI techniques. [ 39 , 48 , 57 , 65 ] …”

“…Two linguistic techniques, syntactic (syntax) and semantic analysis, have assisted machine translation and information retrieval [ 97 ]. Syntactic analysis refers to the grammatical linguistic rules that lead to the well-known subject, action, and object (SAO) structure in S-TRIZ [ 65 ]. Part-of-speech (POS) tagging techniques are primarily used for syntactic analysis.…”

“…For instance, a t -test can be used to assess the average novelty of design ideation in an experiment on a student's mind [ 48 ]. In another experiment, after experts annotated the data, Cohen's kappa coefficient was applied to measure inter-rater reliability as statistic performance [ 65 ]. The performance of clustering results for papers and patent corpuses are evaluated by determining cluster sparsity coefficients presented by ORCLUS [ 49 ].…”