KinderMiner Web: a simple web tool for ranking pairwise associations in biomedical applications

Many important scientific discoveries require lengthy experimental processes of trial and error and could benefit from intelligent prioritization based on deep domain understanding. While exponential growth in the scientific literature makes it difficult to keep current in even a single domain, that same rapid growth in literature also presents an opportunity for automated extraction of knowledge via text mining. We have developed a web application implementation of the KinderMiner algorithm for proposing ranked associations between a list of target terms and a key phrase. Any key phrase and target term list can be used for biomedical inquiry. We built the web application around a text index derived from PubMed. It is the first publicly available implementation of the algorithm, is fast and easy to use, and includes an interactive analysis tool. The KinderMiner web application is a public resource offering scientists a cohesive summary of what is currently known about a particular topic within the literature, and helping them to prioritize experiments around that topic. It performs comparably or better to similar state-of-the-art text mining tools, is more flexible, and can be applied to any biomedical topic of interest. It is also continually improving with quarterly updates to the underlying text index and through response to suggestions from the community. The web application is available at https://www.kinderminer.org.

Latviešu literārā biedrība un skolotāja Pētera Zēvalda savāktais jaunvārdu un mazāk zināmo vārdu krājums (1874)

This research is a part of the analysis of the collection of articles of the Latvian Literary Society (Lettisch-literärische Gesellschaft) Magazin, herausgegeben von der Lettisch-Literärischen Gesellschaft (so-called “Magazin”) with special emphasis to the materials important in the history of Latvian language, especially, of the development of terminology in Latvian. This paper deals with the almost forgotten author Pēteris Zēvalds (Peter Seewald) (ca. 1838/1839–1910), a schoolteacher in Jelgava (Mitau) and active contributor of “Magazin” from 1865 to 1877. In 12 sequels (published in eight issues of “Magazin”), he collected 2198 words and expressions from his native place, now Birzgale parish, originally, from the private manor in Linde parish (aus Privatgut Lindenschen Gemeinde in Kurland)). This paper provides a more detailed analysis of two other word collections (both published in 1874 in issue 3 of XV volume of “Magazin”) of Zēvalds. They contained current lexis, including a lot of new terms. The first one is created as a successive excerpt of neologisms and lesser-known words (in Latvian-German comparison) from the weekly newspaper “Baltijas Vēstnesis” (using numbers 44th, 48th, 49th un 51st from 1872 and successive first 49 from 1873, excl. number 9th). There are 467 numbered entries or 495 words in total. Most of the words in Zēvalds’s collection are related to terminological lexis. They can be divided into four groups (indicating the serial number used in the original): (1) terms currently used in the same form or with minor changes of ending (7. greizsirdība, 29. pilnvare, 41. māksla, 33. cēlons, 69. veicināšana, 88. pirmvaloda, 237. izvilktne (= atvilktne), 275. viels); (2) terms which have a different correspondence in modern Latvian (32. valodas=iztirzāšana (= valodniecība), Sprachforschung; 49. lietuve (= lejkanna), Gieskanne; 54. rakstiens (=raksts, dokuments), Schriftstück; 183. atvēles=zīme (= atļauja), Erlaubnißschein; 187. jūras=pāržmauga (= jūrasšaurums), Meerenge); (3) words, that could still be considered as potential terms (11. tiesas=laulība, Zivilehe; 37. sauljumte, Sonnenschirm) and (4) those, which, from the moment of fixation, can be considered occasional words (43. spīdgans, Sternschnuppen, Meteor; 387. muldu=valsts, Reich der Träume; 409. tulpete (= runas=vieta), Katheder, Rednerbühne). Zēvalds’s collection is a unique material that allows identifying the perception of the interested reader about the lexical neologisms in one newspaper. Zēvalds’s second collection of words contains the technical expressions from legal country house purchase agreements. There are 83 numbered entries (in German-Latvian comparison) from the legal contract written in German with a lot of additional conditions and long pay-outs delay. This list (at least the part containing legal terms and terminological word-groups) can be regarded as the first term bulletin in Latvian. The possible addressee of this term list is the parish pastors to whom farmers sought advice on such matters. Publications of the teacher Pēteris Zēvalds is a small but interesting episode in the history of the Latvian language, which has not earned attention so far but provides researchers with interesting material about the lexical development in the 70s of XIX century.

Automatic Sentiment Detection in Naturalistic Audio

This paper proposed another Audio notion investigation utilizing programmed discourse acknowledgment is an arising research territory where assessment or opinion showed by a speaker is identified from regular sound. It is moderately under-investigated when contrasted with text-based notion identification. Separating speaker estimation from common sound sources is a difficult issue. Nonexclusive techniques for feeling extraction by and large use records from a discourse acknowledgment framework, and interaction the record utilizing text-based estimation classifiers. In this examination, we show that this standard framework is imperfect for sound assessment extraction. Then again, new engineering utilizing watchword spotting (UWS) is proposed for assumption discovery. In the new engineering, a book-based assessment classifier is used to naturally decide the most helpful and discriminative feeling bearing watchword terms, which are then utilized as a term list for UWS. To get a minimal yet discriminative assumption term list, iterative element enhancement for most maximum entropy estimation model is proposed to diminish model intricacy while keeping up powerful grouping precision. The proposed arrangement is assessed on sound acquired from recordings in youtube.com and UT-Opinion corpus. Our exploratory outcomes show that the proposed UWS based framework fundamentally outflanks the conventional engineering in distinguishing assumption for testing reasonable undertakings.

Development of a list of terms in Brazilian Portuguese for the Qualitative Behaviour Assessment of broiler chickens

Qualitative Behaviour Assessment (QBA) is a methodological approach to assess the whole animal using terms to describe and quantify the emotionally expressive qualities of behaviour and identifying larger patterns of expressivity through multi-variate statistical integration. A key condition for the success of QBA is achieving a common understanding of the meaning of descriptive terms by raters. Based on this, our study aimed to develop a list of terms in Brazilian Portuguese for the QBA of broiler chickens (Gallus gallus domesticus), and to test this list by studying its inter- and intra-rater reliability. Fourteen experts participated in a workshop and developed a list of 25 QBA terms, and 40 undergraduates tested this list by scoring 18 video clips using a 125-mm visual analogue scale. Principal Component Analysis was used to analyse observers' scores. Principal Component (PC) 1 ranged from disturbed/frustrated to comfortable/lively, suggesting this PC may be interpreted in terms of emotional valence. PC2 ranged from calm/dull to agitated/active, suggesting this PC indicates the level of arousal/energy of the birds. Both PC1 and PC2 clip scores showed good inter- and intra-rater reliability. This study demonstrates the importance of producing QBA term lists bottom-up as opposed to merely translating pre-existing lists from the scientific literature. Results suggest the standardised Portuguese QBA term list developed in this study is reliable in assessing the expressive qualities of broiler behaviour; therefore, a next step is to test it on-farm with experienced raters and further refine it concerning terms related to poor welfare.

KinderMiner Web: a simple web tool for ranking pairwise associations in biomedical applications

Many important scientific discoveries require lengthy experimental processes of trial and error and could benefit from intelligent prioritization based on deep domain understanding. While exponential growth in the scientific literature makes it difficult to keep current in even a single domain, that same rapid growth in literature also presents an opportunity for automated extraction of knowledge via text mining. We have developed a web application implementation of the KinderMiner algorithm for proposing ranked associations between a list of target terms and a key phrase. Any key phrase and target term list can be used for biomedical inquiry. We built the web application around a text index derived from PubMed. It is the first publicly available implementation of the algorithm, is fast and easy to use, and includes an interactive analysis tool. The KinderMiner web application is a public resource offering scientists a cohesive summary of what is currently known about a particular topic within the literature, and helping them to prioritize experiments around that topic. It performs comparably or better to similar state-of-the-art text mining tools, is more flexible, and can be applied to any biomedical topic of interest. It is also continually improving with quarterly updates to the underlying text index and through response to suggestions from the community. The web application is available at https://www.kinderminer.org.

BEERE: a web server for biomedical entity expansion, ranking and explorations

BEERE (Biomedical Entity Expansion, Ranking and Explorations) is a new web-based data analysis tool to help biomedical researchers characterize any input list of genes/proteins, biomedical terms or their combinations, i.e. ‘biomedical entities’, in the context of existing literature. Specifically, BEERE first aims to help users examine the credibility of known entity-to-entity associative or semantic relationships supported by database or literature references from the user input of a gene/term list. Then, it will help users uncover the relative importance of each entity—a gene or a term—within the user input by computing the ranking scores of all entities. At last, it will help users hypothesize new gene functions or genotype–phenotype associations by an interactive visual interface of constructed global entity relationship network. The output from BEERE includes: a list of the original entities matched with known relationships in databases; any expanded entities that may be generated from the analysis; the ranks and ranking scores reported with statistical significance for each entity; and an interactive graphical display of the gene or term network within data provenance annotations that link to external data sources. The web server is free and open to all users with no login requirement and can be accessed at http://discovery.informatics.uab.edu/beere/.

Association between past medical history (PMH) of autoimmune events and adverse events of special interest (AESI).

2565 Background: PD-1/L1 inhibitor therapy has become the standard of care for many advanced solid tumors. A notable limitation of PD-1/L1 inhibitor therapy is the concern for AESI that are likely to be immune related. It is unclear whether a history of autoimmune events is a predisposing risk making these adverse events more likely. We aimed to evaluate the association between autoimmune-associated PMH and AESI on PD-1/L1 inhibitors. Methods: We pooled data across seven pivotal trials for PD-1/L1 inhibitors in urothelial cancer (UC) identifying patients with AESI submitted from 2016 - 2017. AESI were determined using a pooled preferred term list for each trial. Information collected from trials included PMH, AESI events and toxicity grade. Results: In total, 1747 immunotherapy treated patients were identified, with 1068 (61%) having an AESI and 277 (26%) having an autoimmune-related PMH. The most common autoimmune-associated events in the PMH were thyroid disorders (64%), asthma (23%), atopic dermatitis/eczema (6%), irritable bowel syndrome (5%), and psoriasis (4%). AESI occurred in 68% of patients with an autoimmune-related PMH and 60% of patients without an autoimmune-related PMH. The most common AESI in patients with an autoimmune-related PMH were diarrhea (35%), rash (27%), renal disorder (27%), and pruritis (23%). The most common AESI in the general trial population were diarrhea (28%), pruritis (26%), rash (25%), increased creatinine (14%), and elevated AST and ALT (10% and 9%). The majority of events were Grade 1-2 (87% in patients with an autoimmune-related PMH and 84% in the general trial population). Conclusions: Pooled clinical trial data shows a slight numeric increase in AESIs in patients with autoimmune-associated PMH. Limitations include potential lack of consistency of PMH documentation and adverse event reporting. There did not appear to be a pattern of association between PMH and type of AESI event. Grades of AESI events in the population with autoimmune-associated PMH were similar to the general trial population. This suggests that PD-1/L1 inhibitors may be safely administered to patients with UC and a PMH of some autoimmune-associated events. Further exploration is needed.

Machine Learning

A Bridge Data Dictionary contains an exhaustive list of terms used in the field of bridges. These terms are classified in systems in order to avoid any lacks, to identify all the expected object attributes, and to allow machines to understand the associated concepts. The main objectives of a Bridge Data Dictionary are many: ensure the sustainability of information over time; facilitate information exchange between the actors of the same project; ensure interoperability between the software packages. Other objectives have been reached during the process: to test a working methodology to be applied by other infrastructure domains (Roads, Rails, Tunnels, etc.); to check the current functions and capabilities of a buildingSMART Data Dictionary platform; and to define a common term list, in order to facilitate standardization and IFC-Bridge classes' development.

Nested term recognition driven by word connection strength

Domain corpora are often not very voluminous and even important terms can occur in them not as isolated maximal phrases but only within more complex constructions. Appropriate recognition of nested terms can thus influence the content of the extracted candidate term list and its order. We propose a new method for identifying nested terms based on a combination of two aspects: grammatical correctness and normalised pointwise mutual information (NPMI) counted for all bigrams in a given corpus. NPMI is typically used for recognition of strong word connections, but in our solution we use it to recognise the weakest points to suggest the best place for division of a phrase into two parts. By creating, at most, two nested phrases in each step, we introduce a binary term structure. We test the impact of the proposed method applied, together with the C-value ranking method, to the automatic term recognition task performed on three corpora, two in Polish and one in English.