Learning keyword phrases from query logs of USPTO patent examiners for automatic query scope limitation in patent searching

Professional search in patent repositories poses several unique challenges. One key requirement is to search the entire affected space of concepts, following well-defined procedures to ensure traceability of results obtained. Several techniques have been introduced to enhance query generation, preferably via automated query term expansion, to improve retrieval effectiveness. Currently, these approaches are mostly limited to computing additional query terms from patent documents based on statistical measures. For conceptual search to solve the limitation of traditional keyword search standard dictionaries are used to provide synonyms and keyword phrases for query refinement. Studies show that these are insufficient in such highly specialized domains. In this paper, we present an approach to extract keyword phrases from query logs created during the validation procedure of the patent applications. This creates valuable domain-specific lexical databases for several specific patent classes that can be used to both expand as well as limit the scope of a patent search. This provides a more powerful means to guide a professional searcher through the search process. We evaluate the lexical databases based on real query sessions of patent examiners.     

Blockchain breeding grounds: Asia's advance over the USA and Europe

Blockchain is argued to disrupt almost every industry and research field. Among other technologies, it contributes to the next fourth industrial revolution. While prominent technologies, such as AI and IoT, have already proven their potential and value in many applications, Blockchain applications have gathered only limited appeal. We ask, which countries drive inventive activity in Blockchain and to which magnitude and type. Using worldwide Blockchain patent applications from 2009 to 2020 as inventive outcome indicators the National Breeding Ground (NBG) index and the International Breeding Ground (IBG) index are calculated. The NBG index is related to the interest of a country to file Blockchain patents in its domestic market. The IBG index is related to the interest of a country to file and exploit Blockchain patents in foreign markets. We observe China and the USA as strong national and international breeding grounds. In particular, Asian countries drive Blockchain inventions. European countries contribute marginally to the Blockchain patent landscape and, together with the USA, are losing ground to Asia. Comparisons to prior work on Blockchain patenting and research patterns partially reflect our results. However, comparisons to AI suggest different patenting and exploitation strategies for Blockchain and AI in China.     

Search strategies at the European Patent Office

Prior-art search is a critical step towards determining whether a patent can be granted or not. In 2016, an internal project called Search Workflow Modelling (SWM) was launched at the European Patent Office (EPO) for building a search knowledgebase, which contains a set of models that record not only the current situation on how patent examiners deal with prior-art search (i.e. the as-is models), but also their requirements of being able to do a more efficient and effective search (i.e. the to-be models). We use the Fact-based Modelling (FBM) approach for formalizing search ontologies, which cover a common vocabulary, relations between concepts related to search, and constraints applicable to these relations. We use a hybrid modelling approach of Business Process Modelling Notations and Case Management Model and Notations (BPMN/CMMN) to model search work flows. A patent search strategy typically involves at least one FBM model and at least one BPMN/CMMN model. In this paper, we will illustrate 5 types of existing search strategies (including recursive flow patterns and FBM models for future search features), and future search strategies. The SWM empirical studies in this paper are being put into practice in the ongoing projects concerning search tools at the EPO.     

Legal Validity Metrics for Patent Offices and Law Firms – Recycling of Patent Court Appeal Decisions for Quality Management

In all relevant patent systems an adversely affected party has the possibility to appeal the patent office's decision in front of a patent court or Board of Appeal (BoA). Within a European context, the EPO is, in a way, even engaging in de facto competition with national patent offices. As an example, the German Patent and Trademark Office (DPMA) provides quasi identical patent products and offers the same kind of legal recourse to the German patent court (BPatGer) for these products. Both offices offer an opposition procedure which is open for appeal, and also in the case of a refusal, both offices offer the possibility of appeals. Thus, the EPO Boards of Appeal are engaged in making decisions every year on around 2000 cases and, e.g., the BPatGer on 600 cases regarding the legal validity of the first instance's work. It is thus remarkable that, so far, nobody has systematically included the work of the patent courts in a statistical legal validity evaluation in the context of the Patent & Trademark Offices' (PTOs') quality assurance systems.Since there is obviously a need to establish a common standard for the comparison of first and second instance legal teachings in patent law, a methodology for comparative legal validity analysis is proposed in this paper. I have tested and optimised the approach while working as a project manager at the European Patent Office, based on analysis of 2300 appeal decisions from seven annual batches.     

Detecting Multi Word Terms in patents the same way as entities

In English patent document information retrieval, Multi Word Terms (MWTs) are an important factor in determining how relevant a patent document is for a particular search query. Detecting the correct boundaries for these MWTs is no trivial task and often complicated by the special writing style of the patent domain. In this paper we describe a method for detecting MWTs in patent sentences based on a method for detecting technical entities using deep learning. On our annotated dataset of 22 patents, our method achieved an average precision of 0.75, an average recall of 0.74 and an average F1 score of 0.74. Further, we argue for the use of domain specific word embedding resources and suggest that our model mostly learns whether individual words should be included in MWTs or not.     

Design Patent Map: An Innovative Measure for Corporative Design Strategies

Abstract:

The design patent map is one of the essential strategic measures for formulating design strategies. Because the design patent is considered less definite than other patent categories, it is relatively difficult to create a design patent map. No study has been discovered in a worldwide search of patent documents or publications that is relevant to design patent maps, proving the innovative nature of this study. The authors have extracted 96 representative works for bicycle frame design patents that were published from 1992 to 2003 and have used them as the study samples. To ensure precise interpretation of the technical data of bicycle frames and to create the design patent map, the authors have interviewed 11 design patent examiners who have an average of 10 years of experience in patent examination, and are currently working in the Taiwan Intellectual Property Office. Based on the matrix for the available patenting capability and the degree of need for patents, the authors have further converted the design patent map into nine feasibly specific design strategies: (1) overall seizure, (2) expansion, (3) closing-in, (4) envelopment, (5) dispersion, (6) pincer attack, (7) division, (8) deviation, and (9) abandonment. As a result, enterprises will be able to employ the design patent map developed in this study to customize their own design strategies. Such a strategic measure will soon become a new competitive edge for enterprises with the arrival of the knowledge economy era.     

Identifying author–inventors from Spain: methods and a first insight into results

The purpose of this paper is twofold: methodological and empirical. Methodologically, we describe a matching and disambiguation procedure for the identification of author–inventors (researchers who publish and patent) located in the same country. Our methodology aims to maximize precision and recall rates by taking into account national name writing customs and country-specific dictionaries for person and institution names (academic and non-academic) in the name matching stage and by including a recursive validation step in the person disambiguation stage. An application of this methodology to the identification of Spanish author–inventors is described in detail. Empirically, we present the first results of applying the described methodology to the matching of all SCOPUS 2003–2008 publications of Spanish authors to all 1978–2009 EPO applications with Spanish inventors. Using this data, we identify 4,194 Spanish author–inventors. A first look at their patenting and publication patterns reveals that they make quite a significant contribution to the country’s overall scientific and technological production in the time period considered: 27 % of all EPO patent applications invented in Spain and 15 % of all SCOPUS publications authored in Spain, excluding non-technological disciplines. To our knowledge, this is the first time that a large scale identification of author–inventors from Spain has been done, with no limitation in terms of fields, regions or types of institutions. We also make available online for scientific use an anonymized subset of the database (patent applications invented by authors affiliated to Spanish public universities).     

A guide to efficient keyword, sequence and classification search strategies for biopharmaceutical drug-centric patent landscape searches - A human recombinant insulin patent landscape case study

Searching biopharmaceutical drug-related patent information is generally considered to be challenging. In particular, setting up efficient search strategies for comprehensive retrieval of high amounts of patent documents related to processes and methods of use, that achieve a reasonable level of precision, but still remain within a particular search scope. While it is generally accepted that patent information cannot be searched using standardized approaches, it is desirable to have a basic rule set for successful biopharmaceutical drug-related patent information retrieval, particularly facing a steady flow of patent expirations for prominent biologic drugs. The present human recombinant insulin case study shows an assessment of keyword, sequence and classification search strategies for establishing biopharmaceutical drug-centric patent landscapes. The search results of both crude and sophisticated keyword search strategies, as well as of a sequence search strategy, were compared in terms of the key information retrieval quality indicators; the recall and the precision. Through analyses of the relevant retrieved documents, a quality assessment of keyword choice is provided, as well as determining focused IPC and Derwent Manual classification codes and terminology from original patent and Derwent documentation abstract titles. All of which can be used for setting up more efficient search strategies and facilitated document categorization.     

Defensive Disclosure of Patentable Inventions under Antitrust Enforcement

Why do large firms disclose patentable inventions defensively? We study IBM's disclosure and patenting activity and find that stronger antitrust enforcement and consent decrees limiting patent protection lead to more defensive disclosure. We extend our analysis to Xerox and show, using a differences-in-differences approach, a decline in patent propensity and an increase in patent scope in patent classes affected by the 1975 consent decree. We extend our analysis to include AT&;T, the other exceptionally large firm with defensive-disclosure activity, and show that their patenting propensity declined under increased antitrust enforcement relative to other firms in the industry. Overall, we show how these firms used defensive disclosure as a strategy to balance the benefits of patenting with the costs of uncertain antitrust enforcement. Our results are especially relevant in light of the defensive patent strategies many firms employ to preempt and cope with patent litigation.     

The EPO search documentation in the face of software related inventions

Software related inventions being not simply excluded from patenting — as confirmed by the recent amendments of the EPO Guidelines — the question arises whether DG1 should adapt its search documentation in order to enable such inventions to be properly searched.As a matter of fact, the question concerns only the non-patent literature (NPL), mostly articles from journals and conference proceedings, because these documents are only selectively classified for inclusion in the search documentation — whereas patent documents are all systematically classified and included — and besides, software matters are rarely dealt with in the present patent literature. In view of the foregoing, the paper illustrates the importance of non-patent literature for search purposes and considers the selection criteria for articles and for the collection of technical journals.     

Determinants of patent withdrawals: Evidence from a sample of Italian applications with the EPO

This paper analyses the propensity to withdraw European patent applications within a regional sample of Italian applicants. The procedure for obtaining a granted patent from the EPO is composed of a series of sequential and selective steps imposing additional costs to the applicants. Accordingly, we argue that early withdrawals – i.e. those occurring before the proper examination process begins – should be treated separately from late withdrawals. Our findings show the probability of an early withdrawal is higher for applicants with lower resources and competencies and rises with the number of backward citations added by EPO examiners to the original application. Late withdrawals, instead, are negatively affected by one factor only: the size of patent family, which approximates the sunk costs born by applicants in order to extend the geographical scope of patent protection. Such a limited explanation suggests that the interventions of EPO examiners are likely to play a significant role in inducing late withdrawals.     

Successful European oppositions (part II) Analysis for the patent information professional

European patents can be opposed within nine months from grant. The result of a successful opposition can be a complete revocation of a granted patent. The main reason for revocation is lack of inventive step (47%). From the patent information perspective, an opposition search against an EU-patent needs, therefore, to be first be focused on inventive step. The EPO has clear guidelines on how prior art can be used to attack the inventive step and how it is used during examination. Following the Problem-Solution-Approach used by the EPO, an opposition search should follow the same steps.     

The future of patent information––a user with a view

The future of patent information is considered from the combined perspective of an information specialist, a patent agent, an inventor and a corporate patent information user. A brief overview of current patenting trends is highlighted, followed by a close look at future trends in the patent information field, including changes in the delivery, analysis and management of patent information, and in the role of intermediaries. A few examples are provided in support of forward-looking statements to illustrate the challenging dynamics involved in this field. Finally, a long-term view is considered, which indicates there remains much need for improvement.     

EPOS: A linguistic tool for full text search

EPOS (EPO Synonyms) is a system being devised by the EPO as a full text search tool to assist the examiner in querying databases. If an examiner knows one word expressing his field of interest, EPOS delivers to him all the ECLA (EPO classification) groups where this basic term is considered relevant. The examiner then selects the right words appropriate to the desired search query.     

When is a search not a search? Part 3 – The worst of the worst: When complex applications also lack unity

Complex Applications are often difficult for patent examiners to search comprehensively due to problems relating to Articles 83 and 84 of the EPC. Non-Unity occurs when there is no single general inventive concept in an application (Article 82, EPC). But what does an examiner do when his application contains both? This article explores the particular problems and possible solutions when an application is both complex and lacking unity, and also notes that the streamlined EPO procedure can deal with these more effectively.Whilst the article comments on EPO procedure, it is the opinion of the authors alone and is not an official EPO view.     

Innovative Output,Productivity and Profitability. A Test Comparing USPTO and EPO Data

The aim of this paper is to test whether patent‐based indicators are still reliable measures of innovativeness in light of organizational changes in the field of Intellectual Property Rights (IPR) protection and the regulatory reforms already occurred and under way, respectively, at the US Patent and Trademark Office (USPTO) and the European Patent Office (EPO). For most high‐tech industries, patents represent an outcome of the production process and their number can be taken as a proxy for a firm's ability to improve its productivity growth and profitability. The case study reported here concerns the biotechnology industry in Italy, whose firms, by definition, have Intellectual Property (IP) activities in their portfolios. For this purpose, we use a unique dataset which collects balance sheet items and patent information from EPO and USPTO. After linking firms' financial and production data with the patent information, we estimate a modified knowledge production function in which the dependent variable is alternatively (labor) productivity growth and profitability. Although based on a quite small sample, our findings provide some indication of a statistically significant relationship between patents with the EPO and both productivity growth and, in particular, profitability. This suggests that firms might pursue different strategies when patenting with the USPTO and the EPO.     

Ontology-based Tamil–English cross-lingual information retrieval system

Cross-lingual information retrieval (CLIR) systems facilitate users to query for information in one language and retrieve relevant documents in another language. In general, CLIR systems translate query in source language to target language and retrieve documents in target language based on the keywords present in the translated query. However, the presence of ambiguity in source and translated queries reduces the performance of the system. Ontology can be used to address this problem. The current approaches to ontology-based CLIR systems use manually constructed multilingual ontology, which is expensive. However, many methods exist to automatically construct ontology for any domain in English but not in other languages like Tamil. We propose a methodology for Tamil–English CLIR system by translating the Tamil query to English and retrieve pages in English to address these issues. Our approach uses a word sense disambiguation module to resolve the ambiguity in Tamil query. An automatically constructed ontology in English is used to address the ambiguity of English query. We have developed a morphological analyser for Tamil language, Tamil–English bilingual dictionary and named entity database to translate a Tamil query to English. The translated query is reformulated using ontology and the reformulated queries are given to a search engine to retrieve English documents from the Internet. We have evaluated our methodology for agriculture domain and the evaluation results show that our approach outperforms other approaches in terms of precision.     

Citations as a measure of technological impact: A review of forward citation-based measures

The number of forward citations a patent receives accumulates over time and appears to be correlated to the patent‘s (i.e. invention's) technological impact. A dominant theory suggests that highly cited patents contain an important technological advance. However, a variety of citation based measures have been proposed by different authors. This study, via a narrative literature review, identified nine forward citation-based measures that appear of particular relevance. We describe each measure and present them in a comparative format.The measures are divided into two broad categories: firstly the ones that are particularly relevant to the patent level (citation index, forward citation frequency, generality, influence), and secondly the ones that are relevant to the patent portfolio level (current impact index, herfindal-hirschman index, hindrance index, relative patent position, technology strength).We hope research scholars and industrial users find this review helpful for citation analysis and intellectual property analytics, especially when wanting to employ forward citation-based measures to assess technological impact.     

Measuring patent quality: A claim and search report approach

This paper presents a new prospective metric for assessing the novelty and inventiveness of patents. It does this by using initial patent search reports and examiner's intuition about the impact of adverse citations on patent claim survival. The paper then demonstrates the metric by evaluating the quality of Switzerland's national patent stock using a selection model, finding that between 84 and 90% of the country's national patents would likely not survive examination at the European Patent Office. In doing so, it contributes to the larger literature on patent assessment, underscores the relevance of patent strategy in the observed characteristics of patents, and removes some of the ambiguity in the academic literature about backward citations.