Transformer-Based Patent Novelty Search by Training Claims to Their Own Description

In this paper we present a method to concatenate patent claims to their own description. By applying this method, bidirectional encoder representations from transformers (BERT) train suitable descriptions for claims. Such a trained BERT could be able to identify novelty relevant descriptions for patents. In addition, we introduce a new scoring scheme: relevance score or novelty score to interprete the output of BERT. We test the method on patent applications by training BERT on the first claims of patents and corresponding descriptions. The output is processed according to the relevance score and the results compared with the cited X documents in the search reports. The test shows that BERT score some of the cited X documents as highly relevant.

TriatoScore: an entomological-risk score for Chagas disease vector control-surveillance

Background Triatomine bugs transmit Chagas disease across Latin America, where vector control-surveillance is increasingly decentralized. Locally run systems often deal with highly diverse native-vector faunas—plus, in some areas, domestic populations of non-native species. Flexible entomological-risk indicators that cover native and non-native vectors and can support local decision-making are therefore needed. Methods We present a local-scale entomological-risk score (“TriatoScore”) that leverages and builds upon information on the ecology-behavior and distribution-biogeography of individual triatomine bug species. We illustrate our approach by calculating TriatoScores for the 417 municipalities of Bahia state, Brazil. For this, we (i) listed all triatomine bug species recorded statewide; (ii) derived a “species relevance score” reflecting whether each species is native/non-native and, if native, whether/how often it invades/colonizes dwellings; (iii) mapped each species’ presence by municipality; (iv) for native vectors, weighted presence by the proportion of municipal territory within ecoregions occupied by each species; (v) multiplied “species relevance score” × “weighted presence” to get species-specific “weighted scores”; and (vi) summed “weighted scores” across species to get municipal TriatoScores. Using standardized TriatoScores, we then grouped municipalities into high/moderate/low entomological-risk strata. Results TriatoScores were higher in municipalities dominated by dry-to-semiarid ecoregions than in those dominated by savanna-grassland or, especially, moist-forest ecoregions. Bahia’s native triatomines can maintain high to moderate risk of vector-borne Chagas disease in 318 (76.3%) municipalities. Historical elimination of Triatoma infestans from 125 municipalities reduced TriatoScores by ~ 27% (range, 20–44%); eight municipalities reported T. infestans since Bahia was certified free of Trypanosoma cruzi transmission by this non-native species. Entomological-risk strata based on TriatoScores agreed well with Bahia’s official disease-risk strata, but TriatoScores suggest that the official classification likely underestimates risk in 42 municipalities. Of 152 municipalities failing to report triatomines in 2006–2019, two and 71 had TriatoScores corresponding to, respectively, high and moderate entomological risk. Conclusions TriatoScore can help control-surveillance managers to flexibly assess and stratify the entomological risk of Chagas disease at operationally relevant scales. Integrating eco-epidemiological, demographic, socioeconomic, or operational data (on, e.g., local-scale dwelling-infestation or vector-infection frequencies, land-use change and urbanization, housing conditions, poverty, or the functioning of control-surveillance systems) is also straightforward. TriatoScore may thus become a useful addition to the triatomine bug control-surveillance toolbox. Graphical abstract

A multi-source entity-level sentiment corpus for the financial domain: the FinLin corpus

We introduce FinLin, a novel corpus containing investor reports, company reports, news articles, and microblogs from StockTwits, targeting multiple entities stemming from the automobile industry and covering a 3-month period. FinLin was annotated with a sentiment score and a relevance score in the range [− 1.0, 1.0] and [0.0, 1.0], respectively. The annotations also include the text spans selected for the sentiment, thus, providing additional insight into the annotators’ reasoning. Overall, FinLin aims to complement the current knowledge by providing a novel and publicly available financial sentiment corpus and to foster research on the topic of financial sentiment analysis and potential applications in behavioural science.

An Enhanced Semantic Focused Web Crawler Based on Hybrid String Matching Algorithm

Topic precise crawler is a special purpose web crawler, which downloads appropriate web pages analogous to a particular topic by measuring cosine similarity or semantic similarity score. The cosine based similarity measure displays inaccurate relevance score, if topic term does not directly occur in the web page. The semantic-based similarity measure provides the precise relevance score, even if the synonyms of the given topic occur in the web page. The unavailability of the topic in the ontology produces inaccurate relevance score by the semantic focused crawlers. This paper overcomes these glitches with a hybrid string-matching algorithm by combining the semantic similarity-based measure with the probabilistic similarity-based measure. The experimental results revealed that this algorithm increased the efficiency of the focused web crawlers and achieved better Harvest Rate (HR), Precision (P) and Irrelevance Ratio (IR) than the existing web focused crawlers achieve.

Circular Economy Matchmaking Framework for Future Marketplace Deployment

Online marketplaces enable cooperation between potential stakeholders by supporting offer and demand identification at the secondary raw material markets. The use of marketplaces facilitates communication between supply chain actors operating within the same or different industry sectors and enables detection of ways to close the loop of their products. This research investigated which criteria to use for the circular cooperation matching of companies in the context of a marketplace. These criteria were used for the development of a circular economy (CE) matchmaking framework based on a multi-level approach and relevance scoring between the users. The multi-level approach is based on the following criteria: (i) the compliance with circular economy principles, (ii) the material flows analysis, (iii) the greenhouse gas (GHG) emissions from transport. Based on those aspects a Total Relevance Score (TRS) between the stakeholders is calculated. The Total Relevance Score indicates the possibility for successful circular cooperation between two partners who are willing to close their loops with the best possible match. The logic behind the proposed circular economy matchmaking framework is illustrated by four cases using data collected from companies. Recommendations for further deployment are proposed. The developed framework, by incorporating circular economy principles for the first time within the matching algorithm, provides the opportunity for interested stakeholders for more tailored matching and increases their possibilities of finding a perfect match on the secondary raw materials market in terms of circularity. A similar approach has not yet been reported in the literature. The circular economy matchmaking framework constitutes a theoretical basis of an online marketplace to be developed.

A graph-based feature selection method for learning to rank using spectral clustering for redundancy minimization and biased PageRank for relevance analysis

This paper addresses the feature selection problem in learning to rank (LTR). We propose a graph-based feature selection method, named FS-SCPR, which comprises four steps: (i) use ranking information to assess the similarity between features and construct an undirected feature similarity graph; (ii) apply spectral clustering to cluster features using eigenvectors of matrices extracted from the graph; (iii) utilize biased PageRank to assign a relevance score with respect to the ranking problem to each feature by incorporating each feature?s ranking performance as preference to bias the PageRank computation; and (iv) apply optimization to select the feature from each cluster with both the highest relevance score and most information of the features in the cluster. We also develop a new LTR for information retrieval (IR) approach that first exploits FS-SCPR as a preprocessor to determine discriminative and useful features and then employs Ranking SVM to derive a ranking model with the selected features. An evaluation, conducted using the LETOR benchmark datasets, demonstrated the competitive performance of our approach compared to representative feature selection methods and state-of-the-art LTR methods.

Spectral Optimization of Airborne Multispectral Camera for Land Cover Classification: Automatic Feature Selection and Spectral Band Clustering

Hyperspectral imagery consists of hundreds of contiguous spectral bands. However, most of them are redundant. Thus a subset of well-chosen bands is generally sufficient for a specific problem, enabling to design adapted superspectral sensors dedicated to specific land cover classification. Related both to feature selection and extraction, spectral optimization identifies the most relevant band subset for specific applications, involving a band subset relevance score as well as a method to optimize it. This study first focuses on the choice of such relevance score. Several criteria are compared through both quantitative and qualitative analyses. To have a fair comparison, all tested criteria are compared to classic hyperspectral data sets using the same optimization heuristics: an incremental one to assess the impact of the number of selected bands and a stochastic one to obtain several possible good band subsets and to derive band importance measures out of intermediate good band subsets. Last, a specific approach is proposed to cope with the optimization of bandwidth. It consists in building a hierarchy of groups of adjacent bands, according to a score to decide which adjacent bands must be merged, before band selection is performed at the different levels of this hierarchy.

Current technological problems in the production of synthetic resins and wood-based panels

Проведен опрос специалистов по выявлению актуальных проблем в технологии изготовления древесных плит и синтетических смол для их производства. Исследование проведено методом группового анкетного опроса. Анкета содержала 3 перечня вероятных проблем по разделам: производство синтетических смол, производство древесностружечных плит и производство древесноволокнистых плит. Перечни технологических проблем взяты как итоговые результаты прошлых социологических исследований, а также на основании предложений специалистов. Оценка актуальности проблемы производилась по 5-балльной шкале. По величине средневзвешенного балла определяли уровень актуальности технологической проблемы: чем он выше, тем значимее анализируемая проблема. На вопросы анкеты ответили 47 человек. Эксперты имеют большой опыт, их средний стаж работы в отрасли 22,7 года. Более 70 участников опроса работники предприятий, остальные научные сотрудники, проектировщики и преподаватели. Исследовали 10 проблем по технологии синтетических смол, 12 по технологии древесностружечных плит, 10 по технологии древесноволокнистых плит. Для сравнения приведены данные аналогичных опросов, проведённых в 2005 и 2011 гг. По мнению экспертов наиболее значимыми являются проблемы, связанные с созданием и применением синтетических смол в производстве плит. Это новые низкотоксичные смолы, обеспечивающие, помимо снижения содержания формальдегида в смоле и плитах, необходимые физико-механические свойства плит, низкий расход связующего, высокую скорость прессования, экологические параметры производства. В производстве синтетических смол 2 проблемы относятся к категории высокоактуальных (балл актуальности проблемы в диапазоне от 4,01 до 5,00), 8 проблем к категории повышенной актуальности (балл актуальности от 3,01 до 4,00). Высокоактуальные проблемы: снижение токсичности смол для плит и реакционная способность смолы. К проблемам повышенной актуальности относятся вопросы синтеза новых смол, улучшение показателей смол, качество сырья и экологические вопросы производства смол. Среди 12 анализируемых проблем производства древесностружечных плит 2 относятся к категории высокоактуальных, 4 к категории повышенной актуальности, 6 к категории средней актуальности (балл актуальности от 2,01 до 3,00). Высокоактуальные проблемы: снижение токсичности плит и снижение расхода смолы в производстве плит. Проблемы повышенной актуальности: снижение разбухания плит, интенсификация процесса горячего прессования, снижение расхода древесного сырья, переработка низкокачественной древесины. В производстве древесноволокнистых плит 6 проблем относятся к категории высоко актуальных, 3 к категории повышенной актуальности, 1 к категории средней актуальности. Высокоактуальные проблемы для производства древесноволокнистых плит сухого способа: новые экологически чистые смолы, снижение токсичности плит, снижение расхода смолы. Высокоактуальные проблемы для производства древесноволокнистых плит мокрого способа: снижение токсичности сточных вод, новые экологически чистые смолы, очистка сточных вод. A survey of specialists to identify current problems in the technology of making of wood-based panels and synthetic resins for their production was conducted. The study was conducted by a group questionnaire. The questionnaire contained three lists of likely problems by sections: the production of synthetic resins, the production of particleboards and the production of fiberboards. Lists of technological problems are taken as the final results of past sociological studies, as well as on the basis of experts proposals. The topicality of the problem was assessed on a 5-point scale. The weighted average score determined the level of relevance of the technological problem the higher it is, the more significant the problem analyzed. Forty-seven people answered the questionnaire. Experts have a lot of experience, their average work experience in the industry is 22.7 years. More than 70 of the survey participants are employees of enterprises, the rest are scientists, designers and teachers. We studied 10 problems on synthetic resin technology, 12 on technology of particle boards and 10 on the technology of fiberboards. For comparison, data from similar surveys conducted in 2005 and 2011 are given. According to experts, the most significant problems are ones associated with the creation and use of synthetic resins in the production of plates. These are new low-toxic resins, providing in addition to reducing formaldehyde content in resin and slabs the necessary physical-mechanical properties of plates, low binder consumption, high pressing speed and environmental parameters of production. In the production of synthetic resins two problems are in the category of highly relevant (the problem relevance score in the range from 4.01 to 5.00), 8 problems to the category of increased relevance (relevance score from 3.01 to 4.00). Highly relevant problems: reducing the toxicity of slabs resins and the reactionary ability of resin. Issues of increased relevance include the synthesis of new resins, improved resin sins, the quality of raw materials and environmental issues of resin production. Among the 12 analyzed problems of particleboards production, 2 are classified as highly relevant, 4 are in the category of increased relevance and 6 in the category of average relevance (relevance score from 2.01 to 3.00). Highly relevant problems: reducing the toxicity of plates and reducing the consumption of resin in plate production. Problems of increased relevance: reducing swell of plates, intensifying the process of hot pressing, reducing the consumption of wood raw materials, processing of low-quality wood. In the production of fiberboards 6 problems belong to the category of high-actual, 3 to the category of increased relevance and 1 to the category of medium relevance. Highly relevant problems for the production of fiberboards of dry method: new environmentally friendly resins, reduced toxicity of plates, reduced resin consumption. Highly relevant problems for the production of fiberboards of wet method: reducing the toxicity of wastewater, new environmentally friendly resins, wastewater treatment.