Suchergebnisse

As societies gather increasing amounts of data, interest is growing in technology that allows people to explore and analyse it. The process used, known as data mining, offers many potential benefits. In the pharmaceutical field, for example, it can help medical researchers to discover correlations between illnesses and treatments, as well as improve diagnostics, aid understanding of multifactorial diseases, and repurpose existing drugs. Law enforcement agencies can use data mining to spot patterns in the times and dates that crimes are committed, helping them to allocate resources more effectively and prevent crimes from being committed in first place. But questions remain about the legal framework that governs the use of data mining. In particular, do UK and EU copyright rules and the sui generis database right law impede the use of DM? These questions are important, because evidence suggests that when it comes to data mining, Europe is lagging behind the US and a number of Asian countries. One reason for this is that copyright protection in EU and European Economic Area countries is relatively strong, which has a negative net effect on the use of data mining. This paper considers how the relationship between UK and EU copyright and database law and data mining affects three important industries: the pharmaceutical sector, law enforcement and marketing. It concludes that data mining is prima facie infringing. The legal exception and defences currently available are at best insufficient and at worst not applicable to data mining. Such legal uncertainty is likely to have a chilling effect on data mining in Europe. The paper concludes by considering potential legislative and regulatory solutions to ensure that data miners in Europe are able to achieve the potential benefits that data mining offers. One potential solution is for copyright owners to grant researchers licences to mine their data. In addition to the cost and scalability challenges of obtaining licences, however, there is a risk that a licensing-based ...

[EN] Data sharing is a fundamental process for the advancement of both natural and social sciences. Starting from the idea that computers and the internet have drastically changed the world in the last decades, this paper advocates for the creation of a space where archaeologists from around the world can share information about maritime history and exchange data with colleagues. Following the principles of open access, we argue that raw data publication is necessary and significant for the development and democratization of the discipline. This study explains the fundamental aspects of the Nautical Archaeology Digital Library (NADL) and its efforts to standardize information collection for shipwrecks and related sites, so that scholars can create a community to disseminate both raw data and complete information in the field of maritime archaeology. To achieve this, our purpose is to facilitate the development of common-ground methodology and terminology that promotes an intelligible dialogue within the global community of nautical archaeologists. This paper addresses some considerations on terminology and systematization in scientific disciplines and discusses the theoretical and methodological issues linked to the process of making a template for recording shipwrecks. Furthermore, this article analyses some of the problems related to the standardization of description processes and the necessity to create a flexible system that accounts for data diversity. The third section discusses how science is greatly enhanced by publishing information in open access platforms.Highlights:Standardization of data allows robust comparative and inter-subjective analysis of coastal and maritime projects, shipwrecks and nautical technology.Research is strongly benefited by sharing information underlying publications and raw data generated within a project in open source platforms.Digital databases such as NADL enhance cooperative research, as well as teaching and outreach strategies. ; [ES] Compartir información es un proceso ...

Current data needs in the field of Economics are largely met through readily available open sources from government agencies, international organizations and non-profit organizations like the National Bureau of Economic Research that freely provide full-text of working papers and data. Proprietary data is also important for research in this area, but is often only available through library-licensed databases. When novice economics graduate students independently seek data, where do they look and why? Are library-licensed data sources being used in addition to widely known free web resources?Through a series of focus group interviews, I am investigating the strategies used to find information online by graduate and PhD students in Economics degree programs at New York University. The study hopes to understand what type of information and resources students use to conduct their research to evaluate current library and publicly available resources related to Economics. The study will assess the quality of data found, identify data trends and innovative search strategies by students. The results will enhance the library's outreach and teaching strategies to improve students' research skills in finding reliable data and lead to data-driven collection development that is more closely tied with the information seeking behavior of economics students.

IntroductionThe Discharge Abstract Database (DAD) associates ICD-10-CA diagnosis codes with inpatient care episodes at acute-care facilities. Codes are assigned by human coders, based on chart review. Coding guidelines stipulate mandatory coding of major and fatal conditions but only optional coding of secondary conditions, which results in undercoding for many conditions.
Objectives and ApproachThis research evaluates machine learning approaches for identifying and completing records with missing codes, to improve data quality. The Alberta Hospital DAD for 2013-14 was used in this study. We assumed that the existing ICD-10-CA codes in the DAD are correct, and used them as training examples. Several ML classifiers, including logistic regression and random forest, were used to develop models to assess the coding probability, using existing codes and demographic information. 3300 chart-review records were used as the reference standard. We focused on hypertension-related codes. Validity of raw diagnosis codes in the DAD was used as the baseline.
ResultsA record is deemed to have a missing hypertension diagnosis code if the predicted probability is high, but without the diagnosis codes having been assigned by the coders. In the baseline, the original hypertension codes have high PPV (ranging from 0.902 for the age group 35-54 to 1.000 for the age group 18-34) but low sensitivity (ranging from 0.200 for the age group 18-34 to 0.565 for the age group 75+). The most successful models that we have tested so far have provided improvements of 2-6% in the sensitivity, while maintaining the PPV. More improvement is generally seen for the younger age groups. Initial experiments indicate greater improvements in sensitivity may be possible for other conditions, such as peptic ulcer disease and cerebrovascular disease.
Conclusion/ImplicationsMachine learning approaches can be useful and cost-effective for improving data quality in DAD. While the improvements in sensitivity relative to the baseline are modest at present, further experiments with different models and feature sets are warranted. Experiments with other conditions may also be fruitful.

This paper attempts to find a much-needed balance between data protection rights and trade secret rights on customer information in the European Union framework. Our analysis proposes a "shared management" of secret data between businesses and customers based on an operation of de-contextualization of customer databases. Several rights are in conflict in these two legal domains. For instance, the right to access to personal data and the new proposed right to "data portability" conflict with the interests of trade secret holders. What is even more problematic is that both analyzed legal frameworks are more and more based on a "proprietary" approach to data: they are both a form of abstract "monopoly". In a first step we analyze, in comparison with USA law, when and how the scope of data protection and trade secret protection coincide in practice, according to the proposed EU reforms in the field. As illustrated by literature, balancing rules in these two frameworks are vague and schizophrenic. However, from a literal interpretation of the analyzed it is possible to understand a "favor" for data protection rights. In analyzing the apparent favor to personality rights compared to other (e.g., economic) rights we investigate (both in USA and selected European states) trade secrets in the perspective of personality rights and data protection rights. As a result of our study we propose a change in perspective from the contrast between customers and businesses, to the conflict between customers and businessmen that enables us to verify whether and when personality rights of data subjects affect the above-mentioned personality rights of businessman in practice. the paper proposes to "decontextualize" secret data so that customers can access only data strictly related to their biographical information while trade secret holders can be free not to disclose the output of their data processing (behavior evaluation, forecast, studies on life expectancy, personalized marketing plan, pricing, etc.) if disclosure can adversely affect their interests. In this framework the "proprietary" approach of European laws must be caught as an opportunity, not as an obstacle: we can consider secret data as a "shared good" of customers and businessmen. A multi-level management of data should be based on interests that are common to customers and trade secret holders (secrecy and data updating).

The government's efforts to cope with the economic impact of a pandemic COVID-19 by way of giving social assistance to vulnerable populations were not going well and create new problems in the village. Kotabatu Village, Kotaagung Sub-district as a service partner has experienced various problems related to this, including many people who complain that they do not receive assistance, even though the economy is in bad shape due to the COVID-19 pandemic. After being traced, the data collection system for people who received assistance from the impact of COVID-19 was limited by a fast time. The irregularity of target recipients is also caused by the population data collection information system which still uses a manual system. As a result, it creates several obstacles for partners in the process of managing and processing population data, especially related to data mismatches. Constraints often occur especially during the data collection process Kotabatu villagers by members of the family, people moved, people died, people came, people were born, and the manufacture of letter testimony population. The solution to dealing with the problem of irregularity in targeting aid recipients in Kotabatu Village, by assisting partners in carrying out a data improvement system for COVID-19 beneficiaries, this is important considering the poor data collection system has a negative impact on the community and social assistance recipients affected by COVID-19 in Kotabatu village. The outputs of this service are activity videos, the journal Martabe (Journal of Community Service index sinta 4) which will be published in 2021 and articles that will be presented at the Unila LPPM scientific meeting.

Uganda Bureau of Statistics is a semi-automous body established by an Act of Parliament in 1998, to promote the production of reliable official statistics and ensure the development and maintenance of the National Statistical System (NSS). The Bureau publishes regular surveys on family planning, gender, economic indicators, national accounts, trade, livestock, agriculture to mention but a few and forecasts, including the monthly inflation figures for the country. Outreach on the other hand is an activity of providing services to the populations who might not otherwise have access to those services, the outreach services are focused on meeting the need of people at locations where those in need cannot access them. This paper aims at establishing the outreach services that UBOS provides to the data users, the challenges they face in providing these outreach services to the data users, and the recommendations to improve on the outreach services.

ObjectiveA multi-disciplinary E2A group was established as part of a linked administrative data study examining methamphetamine use in Manitoba. In this presentation we will share our experiences of establishing an E2A, embedding stakeholders in the research process, and outline an iterative engagement plan for the remaining study years.
ApproachThe E2A group is led by two researchers with expertise in patient and public engagement and guided by Pal's (2014) work on policy analysis and activation. Pal emphasizes a multidisciplinary and iterative process as the basis of a more inclusive approach to policy development for complex problems, such as the prevalence of methamphetamine use in Manitoba. Our goal in engaging public rightsholders, service providers and knowledge users in the research is to ensure that their first-hand knowledge and perspectives are reflected in the interpretations of the findings and that analyses address identified complexities in a culturally sensitive and equity-focused way.
ResultsIn the fall of 2020 E2A members were recruited, including persons with lived experience using methamphetamines and their families/care-givers, academics, clinicians, government, and non-government stakeholders. Training was provided on the topics of trauma-informed care, public engagement, effects of colonial and racist institutions, and cultural safety. The first E2A meetings co-developed guiding principles, a vision and mission, as well as provided capacity building around administrative data research. This work was done as a foundation for the next two years, wherein the group will make decisions about key variables (ie: between methamphetamine use and mental health), interpretation of results, knowledge mobilization, and policy development using an iterative self-evaluation process. To date challenges addressed included public health restrictions related to Covid-19 and adapting the research flow to centre lived-experience decision making.
ConclusionThe E2A group is a key component of this study and could serve as a model for other administrative data studies. The group prioritizes stakeholder knowledge, interprets results, flags potential biases in the data research process, and ensures the findings are relevant to the people they are meant to support.

The World Values Survey (WVS) is an international research program devoted to the scientific and academic study of social, political, economic, religious and cultural values of people in the world. The project's goal is to assess which impact values stability or change over time has on the social, political and economic development of countries and societies. The project grew out of the European Values Study and was started in 1981 by its Founder and first President (1981-2013) Professor Ronald Inglehart from the University of Michigan (USA) and his team, and since then has been operating in more than 120 world societies. The main research instrument of the project is a representative comparative social survey which is conducted globally every 5 years. Extensive geographical and thematic scope, free availability of survey data and project findings for broad public turned the WVS into one of the most authoritative and widely-used cross-national surveys in the social sciences. At the moment, WVS is the largest non-commercial cross-national empirical time-series investigation of human beliefs and values ever executed. ; The following changes were included in the Version 1.6.2: Romania 2018: updated variables on education Q275-Q278; Q275R-Q278R; Q275A-Q278A. Turkey 2018:updated variables on education Q275-Q278;Q275R-Q278R; Q275A-Q278A. Cyprus 2018: updated variables N_Region_WVS–Region country-specific; SINTLANGUAGE –Language of interview. Lebanon2018: updated variable N_Region_ISO 3166-2 ; The target population is defined as: individuals aged 18 (16/17 is acceptable in the countries with such voting age) or older (with no upper age limit), regardless of their nationality, citizenship or language, that have been residing in the [country/territory] within private households for the past 6 months prior to the date of beginning of fieldwork (or in the date of the first visit to the household, in case of random-route selection). ; The sampling procedures differ from country to country; probability sample: Multistage Sample, Probability Sample, Simple Random SampleRepresentative single stage or multi-stage sampling of the adult population of the country 18 (16) years old and older is used for the WVS 2017-2021. Sample size was set as effective sample size: 1200 for countries with population over 2 million, 1000 for countries with population less than 2 million. Countries with great population size and diversity (e.g. India, China, USA, Russia, Brazil etc.) are requirred to reach an effective sample of N=1500 or larger. Only 2 countries (Argentina, Chile) deviated from the guidelines and planned with an effective sample size below the set threshold. Sample design and other relevant information about sampling were reviewed by the WVS Scientific Advisory Committee and approved prior to contracting of fieldwork agency or starting of data collection. The sampling was documented using the Survey Design Form delivered by the national teams which included the description of the sampling frame and each sampling stage as well as the calculation of the planned gross and net sample size to achieve the required effective sample. Additionally, it included the analytical description of the inclusion probabilities of the sampling design that are used to calculate design weights. ; Interview ; Mode of collection: mixed mode Face-to-face interview: CAPI (Computer Assisted Personal Interview) Face-to-face interview: PAPI (Paper and Pencil Interview) Telephone interview: CATI (Computer Assisted Telephone Interview) Self-administered questionnaire: CAWI (Computer-Assisted Web Interview) Self-administered questionnaire: PaperIn all countries, fieldwork was conducted on the basis of detailed and uniform instructions prepared by the WVS scientific advisory committee and WVSA secretariat. The main data collection mode in WVS 2017-2021 is face to face (interviewer-administered). Several countries employed mixed-mode approach to data collection: USA (CAWI; CATI); Australia and Japan (CAWI; postal survey); Hong Kong SAR (PAPI; CAWI); Malaysia (CAWI; PAPI). The WVS Master Questionnaire was provided in English and each national survey team had to ensure that the questionnaire was translated into all the languages spoken by 15% or more of the population in the country. A central team monitored the translation process.

Over the last two decades many sub-Saharan African countries have devolved rights and obligations in rural natural resource management from state to local communities in an effort to foster social-ecological sustainability and economic development at the same time. Often these governmental projects were launched in settings in which traditional commons, informed by both the demands of traditional subsistence-orientated agrarian systems and the tenure policies of colonial and postcolonial states were well established, and in which power struggles between rivaling traditional authorities, between seniors and juniors, and between state agents and local communities were pertinent. These moves were also embedded in (partially contradictory) discourses on decentralization, political participation, economic empowerment, and neo-liberally inspired commoditization of natural resources. In the process of devolvement rights and obligations were handed over to communities which were formalized in the process: formal membership, social and spatial boundaries, elected leadership, established models of governance, and accountability both to the wider community and to state bureaucracy. New commons1 were established around specified resources: pastures, water, forests, game. In the process these resources were (partially) commoditized: game owned by the community could be sold as trophies for hunting, lands could be rented out to private investors, and water had to be paid for. This contribution is intended to shed light on the process of establishing new commons - in the local context named conservancies - of game management in north-western Namibia. Game on communal lands had been state-owned and state-controlled in the colonial past. This did not preclude poaching but certainly inhibited significant degrees of commoditization. The new commons of game management are meant to do exactly this, in two steps: first specific rights (in this case management rights and transfer rights) are devolved to a well-defined community; then this community (or its committee) decides how to put the newly gained rights to good use and transfers such rights to private investors, tourism entrepreneurs and commercial hunters. While the first step is informed by discourses on participation and co-management, the second step is market-oriented and seeks so-called public-private partnerships.

Cluster analysis has for long been an effective tool for analysing data. Thus, several disciplines, such as marketing, psychology and computer sciences, just to mention a few, did take advantage from its contribution over time. Traditionally, this kind of algorithm concentrates only on numerical or categorical data at a time. In this work, instead, we analyse a dataset composed of mixed data, namely both numerical than categorical ones. More precisely, we focus on profiling visitors of the National Park of Majella in the Abruzzo region of Italy, which observations are characterized by variables such as gender, age, profession, expectations and satisfaction rate on park services. Applying a standard clustering procedure would be wholly inappropriate in this case. Therefore, we hereby propose an unsupervised classification of mixed data, a specific procedure capable of processing both numerical than categorical variables simultaneously, releasing truly precious information. In conclusion, our application therefore emphasizes how cluster analysis for mixed data can lead to discover particularly informative patterns, allowing to lay the groundwork for an accurate customers profiling, starting point for a detailed marketing analysis.

Algorithms are not to be regarded as a technical structure but as a social phenomenon - they embed themselves, currently still very subtle, into our political and social system. Algorithms shape human behavior on various levels: they influence not only the aesthetic reception of the world but also the well-being and social interaction of their users. They act and intervene in a political and social context. As algorithms influence individual behavior in these social and political situations, their power should be the subject of critical discourse - or even lead to active disobedience and to the need for appropriate tools and methods which can be used to break the algorithmic power.

Algorithms are not to be regarded as a technical structure but as a social phenomenon - they embed themselves, currently still very subtle, into our political and social system. Algorithms shape human behavior on various levels: they influence not only the aesthetic reception of the world but also the well-being and social interaction of their users. They act and intervene in a political and social context. As algorithms influence individual behavior in these social and political situations, their power should be the subject of critical discourse - or even lead to active disobedience and to the need for appropriate tools and methods which can be used to break the algorithmic power.