2020 7th Swiss Conference on Data Science (SDS)最新文献

Assessing the democratic legitimacy of public decisions based on Machine Learning algorithms 基于机器学习算法评估公共决策的民主合法性

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00017

María Carolina Jiménez

引用次数: 0

SDS 2020 Commentary SDS 2020评论

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/sds49233.2020.00007

引用次数: 0

Effects of Graph Pooling Layers on Classification with Graph Neural Networks 图池化层对图神经网络分类的影响

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00021

Linda Studer, Jannis Wallau, R. Ingold, Andreas Fischer

引用次数: 1

A Comparative Assessment and Synthesis of Twenty Ethics Codes on AI and Big Data 20项人工智能与大数据伦理规范的比较评估与综合

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00015

M. Loi, Christoph Heitz, M. Christen

引用次数: 11

SDS 2020 TOC SDS 2020 技术选择委员会

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/sds49233.2020.00004

Yasamin Eslahi, Jannis Wallau

引用次数: 0

Cobra: A CLI Tool To Create And Share Reproducible Projects Cobra:用于创建和共享可复制项目的CLI工具

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00016

Lars Vögtlin, Vinaychandran Pondenkandath, R. Ingold

引用次数: 3

Automated Smartification of Notices to Airmen 自动向飞行员发出通知书

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00018

M. Bravin, Daniel Pfäffli, Sita Mazumder, M. Pouly

{"title":"Automated Smartification of Notices to Airmen","authors":"M. Bravin, Daniel Pfäffli, Sita Mazumder, M. Pouly","doi":"10.1109/SDS49233.2020.00018","DOIUrl":"https://doi.org/10.1109/SDS49233.2020.00018","url":null,"abstract":"Notices to airmen (NOTAMs) are text messages targeted to pilots to make them aware of short term events and obstacles. Typical examples are military actions in airspaces or closed runways. NOTAMs are written in a strongly compressed format and vocabulary, making the information hard to access for non-trained readers. Smart NOTAM is a commercial service offered by Skyguide that filters irrelevant messages for a specific route or period. Additionally, it converts the messages into a standardized phraseology and removes irrelevant information. Having access to thousands of NOTAMs together with their smartified version, a transformer model was trained and evaluated to automate the currently manual smartification process. NOTAMs contain a large number of factual details such as coordinates or radio frequencies, making the decoding process vulnerable to the inaccurate generation of such details. Therefore, the model was extended by a pointer-network that allows copying words directly from the source text. It achieves a BLEU score above 86% on unseen test data. Visual inspection of fails reveals many ground truth messages written in other languages than English or containing factual details added from other sources.","PeriodicalId":136287,"journal":{"name":"2020 7th Swiss Conference on Data Science (SDS)","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123463724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

DARE2HACK: Crowdsourcing ideas through hackathons to codesign new human-centric services DARE2HACK:通过黑客马拉松众包创意，共同设计新的以人为本的服务

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00009

M. Flores, M. Golob, Doroteja Maklin, Christopher L. Tucci, S. West, O. Stoll

{"title":"DARE2HACK: Crowdsourcing ideas through hackathons to codesign new human-centric services","authors":"M. Flores, M. Golob, Doroteja Maklin, Christopher L. Tucci, S. West, O. Stoll","doi":"10.1109/SDS49233.2020.00009","DOIUrl":"https://doi.org/10.1109/SDS49233.2020.00009","url":null,"abstract":"Organizations are continuously looking for ways to transform by adopting new practices and technologies in order to become more competitive. They target business growth by increasing their efficiencies, by improving their internal operations and/or delivering novel value-added products and services. This is not easy to accomplish, as it requires defining a new strategy, changing the way of working, training and engaging employees and interacting with the customers to address different global trends.One major challenge companies are facing is how to shift to collaboratively designing new services enabled by digital technologies, integrating talent from an ecosystem of experts. To design new services, organizations can start by reflecting on how they currently execute their innovation process. This latter is all about transforming ideas, data and knowledge into products or services launched successfully to the market. To accelerate their performance, organizations have been changing their practices, from working exclusively in-house to openly collaborating with external talent. Crowdsourcing refers to the necessary activities to collect “out-of-the-box” ideas in order to benefit from the collective brainpower of diverse entrepreneurial individuals and/or their own employees. A Hackathon could be defined as a crowd-sourcing contest, in which a diverse talented crowd is asked to solve a business challenge within 1 to 2 days and only the winners receive compensation or an award. Although hackathons originally focused on coding, lately organizations have been increasingly organizing them to solve business challenges while applying customer-centric approaches such as Design Thinking. This paper provides the foundations to link these approaches and provide insights about a hackathon that enabled innovation experts from different organizations to provide their ideas and identify key services to be delivered by an innovation hub.","PeriodicalId":136287,"journal":{"name":"2020 7th Swiss Conference on Data Science (SDS)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121493618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Uncertainty with deep learning: a practical view on out of distribution detection 深度学习的不确定性:非分布检测的实用观点

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00025

Daniel Siegismund, S. Heyse, Stephan Steigele

引用次数: 2

Intelligent Branding for Less Well-Known Enterprises by Matching Culture Values of Employers and Candidates 通过匹配雇主和候选人的文化价值观，为知名度较低的企业打造智能品牌

2020 7th Swiss Conference on Data Science (SDS) Pub Date : 2020-06-01 DOI: 10.1109/SDS49233.2020.00010

G. Lu, Christian Dollfus, T. Wozniak, Matthes Fleck

{"title":"Intelligent Branding for Less Well-Known Enterprises by Matching Culture Values of Employers and Candidates","authors":"G. Lu, Christian Dollfus, T. Wozniak, Matthes Fleck","doi":"10.1109/SDS49233.2020.00010","DOIUrl":"https://doi.org/10.1109/SDS49233.2020.00010","url":null,"abstract":"This paper develops and tests a novel employer branding toolset by applying state-of-the-art data science and machine learning techniques to a unique dataset on candidates’ culture preferences. The aim of this toolset is to intelligently brand less well-known enterprises by matching the culture values of employers and candidates. In this context, the culture correlation tool compares and ranks the culture preferences of candidates in order to correlate them for an optimized employer branding. The employer clustering tool represents a grouping mechanism, which joins similar companies and organizations regarding their intrinsic culture characteristics, and matches candidates and employers having equivalent culture values. The culture recommender refers to a machine learning based model performing culture recommendations to candidates leveraging selectable attributes, i.e. “Netflix-style” recommendation of employers to candidates. The job predictor predicts how soon a job opening will be filled within our culture recommendation framework, indicating market trends for, e.g., industry sectors and companies. The findings of this study enhance the market competitiveness of less well-known enterprises in acquiring talents with suited culture background.","PeriodicalId":136287,"journal":{"name":"2020 7th Swiss Conference on Data Science (SDS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124714561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1