MLLP-VRAIN Spanish ASR Systems for the Albayzín-RTVE 2020 Speech-To-Text Challenge: Extension

This paper describes the automatic speech recognition (ASR) systems built by the MLLP-VRAIN research group of Universitat Politècnica de València for the Albayzín-RTVE 2020 Speech-to-Text Challenge, and includes an extension of the work consisting of building and evaluating equivalent systems under the closed data conditions from the 2018 challenge. The primary system (p-streaming_1500ms_nlt) was a hybrid ASR system using streaming one-pass decoding with a context window of 1.5 seconds. This system achieved 16.0% WER on the test-2020 set. We also submitted three contrastive systems. From these, we highlight the system c2-streaming_600ms_t which, following a similar configuration as the primary system with a smaller context window of 0.6 s, scored 16.9% WER points on the same test set, with a measured empirical latency of 0.81 ± 0.09 s (mean ± stdev). That is, we obtained state-of-the-art latencies for high-quality automatic live captioning with a small WER degradation of 6% relative. As an extension, the equivalent closed-condition systems obtained 23.3% WER and 23.5% WER, respectively. When evaluated with an unconstrained language model, we obtained 19.9% WER and 20.4% WER; i.e., not far behind the top-performing systems with only 5% of the full acoustic data and with the extra ability of being streaming-capable. Indeed, all of these streaming systems could be put into production environments for automatic captioning of live media streams.

A descriptive analysis of the data availability statements accompanying medRxiv preprints and a comparison with their published counterparts

Objective To determine whether medRxiv data availability statements describe open or closed data—that is, whether the data used in the study is openly available without restriction—and to examine if this changes on publication based on journal data-sharing policy. Additionally, to examine whether data availability statements are sufficient to capture code availability declarations. Design Observational study, following a pre-registered protocol, of preprints posted on the medRxiv repository between 25th June 2019 and 1st May 2020 and their published counterparts. Main outcome measures Distribution of preprinted data availability statements across nine categories, determined by a prespecified classification system. Change in the percentage of data availability statements describing open data between the preprinted and published versions of the same record, stratified by journal sharing policy. Number of code availability declarations reported in the full-text preprint which were not captured in the corresponding data availability statement. Results 3938 medRxiv preprints with an applicable data availability statement were included in our sample, of which 911 (23.1%) were categorized as describing open data. 379 (9.6%) preprints were subsequently published, and of these published articles, only 155 contained an applicable data availability statement. Similar to the preprint stage, a minority (59 (38.1%)) of these published data availability statements described open data. Of the 151 records eligible for the comparison between preprinted and published stages, 57 (37.7%) were published in journals which mandated open data sharing. Data availability statements more frequently described open data on publication when the journal mandated data sharing (open at preprint: 33.3%, open at publication: 61.4%) compared to when the journal did not mandate data sharing (open at preprint: 20.2%, open at publication: 22.3%). Conclusion Requiring that authors submit a data availability statement is a good first step, but is insufficient to ensure data availability. Strict editorial policies that mandate data sharing (where appropriate) as a condition of publication appear to be effective in making research data available. We would strongly encourage all journal editors to examine whether their data availability policies are sufficiently stringent and consistently enforced.

The influence of digitalization on Chinese banks: new financial technologies in a plan economy

Nowadays, digitalization has become one of the most significant global development trends. The Chinese banking system, which has only recently been subjected to liberal reforms, is poorly researched in the viewpoint of the digitalization and financial technology impact on it. In this context, it is interesting to identify the main digitalization features in the PRC banking system and to identify this system’s elements which are most susceptible to digitalization influence. The authors set the task to identify how digitalization changes the PRC banking system, in particular, how it affects shadow banking. The major challenges of the article are the high share of the closed data on the Chinese financial sector, the unclarity of the shadow banking activities in the country and their overall negative influence on the economy of PRC and Asian economies and the highly theoretical nature of the digitalization research of banking in the scientific community. The main article’s results consist in that the main differences between digitalization and fintech are determined; it was also proved the Chinese banking system is more adaptable to changing conditions, and the authors suggested the main measures to reduce the shadow banking’ share in China. The main contribution of the article is the revelation of the possible solutions to the problems of shadow banking in China through the use of the new financial instruments of the digital era.

A descriptive analysis of the data availability statements accompanying medRxiv preprints and a comparison with their published counterparts

ObjectiveTo determine whether medRxiv data availability statements describe open or closed data - that is, whether the data used in the study is openly available without restriction - and to examine if this changes on publication based on journal data sharing policy. Additionally, to examine whether data availability statements are sufficient to capture code availability declarations.DesignObservational study, following a pre-registered protocol, of preprints posted on the medRxiv repository between 25th June 2019 and 1st May 2020 and their published counterparts.Main outcome measuresDistribution of preprinted data availability statements across eight categories, determined by a prespecified classification system.Change in the percentage of data availability statements describing open data between the preprinted and published versions of the same record, stratified by journal sharing policy.Number of code availability declarations reported in the full-text preprint which were not captured in the corresponding data availability statement.Results4101 medRxiv preprints were included in our sample, of which 911 (22.2%) were categorized as describing open data, 3027 (73.8%) as describing closed data, 163 (4.0%) as not applicable (e.g. editorial, protocol). 379 (9.2%) preprints were subsequently published, and of these published articles, only 159 (42.0%) contained a data availability statement. Similar to the preprint stage, most published data availability statements described closed data (59 (37.1%) open, 96 (60.4%) closed, 4 (2.5%) not applicable).Of the 151 records eligible for the comparison between preprinted and published stages, 57 (37.7%) were published in journals which mandated open data sharing. Data availability statements more frequently described open data on publication when the journal mandated data sharing (open at preprint: 33.3%, open at publication: 61.4%) compared to when the journal did not mandate data sharing (open at preprint: 20.2%, open at publication: 22.3%).ConclusionRequiring that authors submit a data availability statement is a good first step, but is insufficient to ensure data availability. Strict editorial policies that require data sharing (where appropriate) as a condition of publication appear to be effective in making research data available. We would strongly encourage all journal editors to examine whether their data availability policies are sufficiently stringent and consistently enforced.

Inequality in the Distribution of Healthcare Human Resources in Poland

Human resources are the major input in health systems. Therefore, their equitable distribution remains critical in making progress towards the goal of sustainable development. The purpose of this study is to evaluate equity in the distribution of healthcare human resources across regions of Poland from 2010 to 2017. This research by applying specifically to Polish conditions will allow the existing gap in the literature to be closed. Data were derived from the Database of Statistics Poland, and the Lorenz Curve/Gini coefficient was engaged as well as the Theil index to measure the extent and drivers of inequality in the distribution of healthcare human resources in macro-regions. Population size along with crude death rates are employed as proxies for healthcare need/demand. This research has several major findings. Mainly, it was found, that the geographical distribution of all types of human resources is less equitable than is the case with population distribution. Relatively lower equity in the access to oncologists, family doctors, and cardiologists was found. There are some noticeable differences between macro-regions in the equity level of healthcare human resources distribution. This research provides various implications for policy and practice and will allow for improved planning and more efficient use of these resources.

Designing Pro-Competitive Research Data Pools: Which EU Competition Remedies for Research Data Silos in Digital Markets?

The study moves from the assumption that the sharing of data can – under specific circumstances – give rise to anticompetitive aggregations of research-valuable data in the form of closed data silos. It addresses the question whether and how competition remedies available under EU law can be used for the design of pro-competitive data pools in digital markets. Interesting suggestions for these purposes are given by the recent enforcement policies enacted by the European Commission in high technology innovation markets. Although aimed at restoring very different anticompetitive conducts, these remedies nonetheless appear to share the common function of opening up established innovation alliances for the transfer of research-valuable information assets to external competing parties. Against this backdrop, the suitability of such information-based remedies in the context of digital markets is questioned. The study ultimately puts forward the opportunity of a close collaboration between competition and data protection authorities for a joint governance of data sharing remedies.

ANALISIS POTENSI KEBERLANJUTAN MULTI USAHA SUB SEKTOR PERIKANAN DI KABUPATEN SIDOARJO

Permasalahan sektor perikanan saling terkait antar sub sektor perikanan tangkap, budidaya, serta olahan dan pemasaran hasil perikanan. Penelitian bertujuan menganalisis potensi keberlanjutan usaha multi sub sektor perikanan skala kecil - menengah oleh masyarakat di Kabupaten Sidoarjo. Penelitian deskriptif dilaksanakan April 2015 s/d April 2018. Survey terestris dengan teknik rekam data tertutup dan terarah menggunakan alat bantu kuesioner. Data diberi bobot dan dianalisa menggunakan analisis shift share dan statistik untuk mengetahui keterkaitan antar sub sektor perikanan terhadap potensi keberlanjutannya. Hasil analisis keberlanjutan usaha tangkap (kode 01.T) bernilai terendah 2,3529 gap 6,0 dari nilai tertinggi 8,3529. Nilai regresi usaha penangkapan ikan Y = 0,005 + 0,961 X menunjukkan usaha penangkapan ikan belum mampu memberdayakan sektor lain. Tingkat signifikansi uji T tidak nyata 22,2%. Nilai R Square 0,005 dan Adjusted R Square -0,061 merepresentasikan tingkat kepercayaan usaha penangkapan ikan sangat rendah. Keberlanjutan usaha perikanan budidaya di tambak (simbol 02.Y) bernilai terendah 2,9783. Nilai regresi linier sebesar Y = 0,980 + 3,375 X menunjukkan usaha budidaya memberikan keberdayaan bagi sub sektor lain secara signifikan 97,8%*. Nilai R Square 0,225 dan Adjusted R Square 0,188 merepresentasikan keberlanjutan usaha budidaya kurang menjanjikan. Keberlanjutan olahan dan pemasaran hasil perikanan (kode 03.U) bernilai terendah 7,2600 dengan shift share gap positif 0,2600. Nilai regresi linier Y = 6,031 + 3,235 X signifikansi 100% menunjukkan usaha olahan dan pemasaran berpengaruh terhadap usaha lainnya, dengan nilai R Square 0,651 dan nilai Adjusted R Square 0,636. Hasil penelitian menyimpulkan bahwa sub sektor olahan dan pemasaran hasil perikanan berpeluang besar untuk ditumbuh-kembangkan.Title: Analysis of Potential Sustainability of Multi Fisheries Sub Sector Business in the Sidoarjo RegencyThe problems of fisheries sector are interrelated between the capture fisheries, cultivation, processing and marketing of fishery products. The research aims to analyze the potential sustainability of small and medium scale multi sub-sector fisheries businesses by people in Sidoarjo Regency. Descriptive research was conducted from April 2015 to April 2018. Terrestrial survey with closed data recording techniques and questionnaires were used in the study. Data were measured and analyzed using shift share matrix and statistics to find out the relation between fisheries sub-sectors to their potential sustainability. Results of the capture business sustainability analysis (code 01.T) have the lowest value of 2.3529 gap 6.0 from the highest value of 8.3529. The regression value of fishing business Y = 0.005 + 0.961 X indicates that fishing businesses have not been able to empower other sectors. The significance level of the T test is not real 22.2%. The R Square value of 0.005 and Adjusted R Square -0.061 represents the relatively low level of trust in fishing businesses. The sustainability of aquaculture business in the pond (symbol 02.Y) has the lowest value of 2.9783. The linear regression value of Y = 0.980 + 3.375 X indicates that cultivation provides empowerment for other sub-sectors significantly of 97.8%*. The value of R Square 0.225 and Adjusted R Square 0.188 representing the sustainability of aquaculture is less promising. Sustainability of processed and marketing of fishery products (code 03.U) has the lowest value of 7.2600 with a positive shift share gap of 0.2600. The linear regression value Y = 6.031 + 3.235 X 100% significance indicates that the processed business and marketing affect other businesses, with the value of R Square 0.651 and the value of Adjusted R Square 0.636. It is concluded that the processed and marketing of fishery products subsector have a great opportunity to be developed.

Towards open data blockchain analytics: a Bitcoin perspective

Bitcoin is the first implementation of a technology that has become known as a ‘public permissionless’ blockchain. Such systems allow public read/write access to an append-only blockchain database without the need for any mediating central authority. Instead, they guarantee access, security and protocol conformity through an elegant combination of cryptographic assurances and game theoretic economic incentives. Not until the advent of the Bitcoin blockchain has such a trusted, transparent, comprehensive and granular dataset of digital economic behaviours been available for public network analysis. In this article, by translating the cumbersome binary data structure of the Bitcoin blockchain into a high fidelity graph model, we demonstrate through various analyses the often overlooked social and econometric benefits of employing such a novel open data architecture. Specifically, we show: (i) how repeated patterns of transaction behaviours can be revealed to link user activity across the blockchain; (ii) how newly mined bitcoin can be associated to demonstrate individual accumulations of wealth; (iii) through application of the naïve quantity theory of money that Bitcoin's disinflationary properties can be revealed and measured; and (iv) how the user community can develop coordinated defences against repeated denial of service attacks on the network. Such public analyses of this open data are exemplary benefits unavailable to the closed data models of the ‘private permissioned’ distributed ledger architectures currently dominating enterprise-level blockchain development owing to existing issues of scalability, confidentiality and governance.