6 |
Ensemble of Deep Masked Language Models for Effective Named Entity Recognition in Health and Life Science Corpora
|
|
|
|
In: ISSN: 2504-0537 ; Frontiers in research metrics and analytics, Vol. 6 (2021) P. 689803 (2021)
|
|
BASE
|
|
Show details
|
|
7 |
Linguistic Metrics for Patent Disclosure: Evidence from University versus Corporate Patents
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Pivot machine translation using Chinese as pivot language
|
|
|
|
In: Liu, Chao-Hong orcid:0000-0002-1235-6026 , Cruz Silva, Catarina, Wang, Longyue orcid:0000-0002-9062-6183 and Way, Andy orcid:0000-0001-5736-5930 (2019) Pivot machine translation using Chinese as pivot language. In: 14th China Workshop, CWMT 2018, 25-26 Oct 2018, Wuyishan, China. (2019)
|
|
BASE
|
|
Show details
|
|
9 |
Contentions of language: Multilingalism and the language regime of the European unitary patent system
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Improving multi word term detection in the patent domain with deep learning ...
|
|
|
|
Abstract:
Im Bereich der Patentsuche spielen technische Begriffe eine wichtige Rolle, wenn es darum geht herauszufinden, welche Patentdokumente einem bestimmten Patentdokument am ähnlichsten sind. Im Englischen bestehen technische Begriffe zumeist aus mehreren Wörtern und formen weiters Substanivgruppen (noun phrases). Letzteres wird von Methoden, die solche Mehrwortbegriffe (multi word terms) erkennen sollen, genutzt. Da aber Patenttexte sich oftmals durch eine Schreibstil kennzeichnen, der von anderen englischen Texten abweicht und Sätze sehr lang werden können, sind übliche Methoden zur Erkennung von linguistischen Informationen weniger effektiv. Weiters werden manche Mehrwortbegriffe eher selten in Patenten verwendet, was Methoden, die die Erkennung von Substantivgruppen sowie Information über die Vorkommenshäufigkeit der Begriffe benötigen, nicht unproblematisch macht. In dieser Arbeit präsentieren wir eine Methode zur Erkennung von Mehrwortbegriffen jeglicher Vorkommenshäufigkeit, die nicht die vorherige ... : In patent document information retrieval, the technical terms that are used in a particular patent document are an important factor in determining what the most relevant related documents are. In English, technical terms often consist of multiple words. Further, the fact that they are mostly noun phrases (NP) is utilized by methods detecting such multi word terms (MWT). However, due to the special nature of the patent domain, such as a special writing style and high maximum sentence length, common methods for extracting linguistic information are less effective. Further, some MWTs can occur very infrequently in patents, which makes the use of methods relying on NP extraction and frequency based information problematic. In this thesis we present a method for detecting even rare MWTs in patent texts that does not require the prior detection of NPs. Using supervised machine learning and an artificial neural network inspired by named entity extraction and keyphrase detection methods, we train models on sentences ...
|
|
Keyword:
Natural language processing; patent text mining
|
|
URL: https://repositum.tuwien.at/handle/20.500.12708/7292 https://dx.doi.org/10.34726/hss.2018.40265
|
|
BASE
|
|
Hide details
|
|
11 |
Dataset used to establish a new model of patent database interpretation for user-centred design ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
Patentes farmacêuticas e inovação : a função moderadora do licenciamento compulsório
|
|
|
|
BASE
|
|
Show details
|
|
13 |
A Hierarchical Feature Extraction Model for Multi-Label Mechanical Patent Classification
|
|
|
|
In: Faculty Publications (2018)
|
|
BASE
|
|
Show details
|
|
16 |
Mining query logs to enhance patent searching ... : Querlyoganalyse zur Verbesserung der Patentsuche ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Perverse Innovation
|
|
|
|
In: William & Mary Law Review, vol 58 (2016)
|
|
BASE
|
|
Show details
|
|
18 |
Criteria-based patent mapping for assessing potential conflicts between patent claims
|
|
Li, Zheng. - : Brunel University London, 2014
|
|
BASE
|
|
Show details
|
|
19 |
How to kill inventors: testing the Massacrator© algorithm for inventor disambiguation ; : Scientometrics
|
|
|
|
In: ISSN: 0138-9130 ; EISSN: 1588-2861 ; Scientometrics ; https://hal.archives-ouvertes.fr/hal-03122461 ; Scientometrics, Springer Verlag, In press, 101 (1), pp.477-504 (2014)
|
|
BASE
|
|
Show details
|
|
20 |
Retroactivity at the Federal Circuit
|
|
|
|
In: Indiana Law Journal (2014)
|
|
BASE
|
|
Show details
|
|
|
|