공유자료 HOME > 자료실 > 공유자료
 
[MIS세미나] AI비즈니스 사례 기사 모음(2019.8~) Part II
관리자 19-09-26 11:28 2,511
52  Developers at Google have created a new framework to build more accurate models for machine vision, language translation, and predictive analytics.  https://www.datanami.com/2019/09/04/google-adds-structured-signals-to-model-training/
53 A team from the University of Oxford developed a facial recognition algorithm that detects, tracks, and recognizes chimpanzees in video footage.  https://www.newscientist.com/article/2215359-ai-facial-recognition-software-now-works-for-wild-chimpanzees-too/
54  Dr. Suraj Kapa, a cardiac electrophysiologist at the Mayo Clinic in Minnesota, speaks with the DocTalk Podcast about his recent study, in which he used an AI to predict a person's health based on only ECG readings.  https://www.mdmag.com/medical-news/doctalk-podcast-using-ai-to-read-ecgs-with-dr-kapa
55 A new AI that combines NL and 3D pose models can translate words into human-like movement. https://venturebeat.com/2019/09/11/ai-researchers-translate-language-into-physical-movement/
56  Google used AI to develop a new report about how organizations can use AI to address looming problems for humanity. https://venturebeat.com/2019/09/12/google-report-assesses-how-ai-can-best-serve-humanity/
57 Machine learning algorithms were able to detect the severity of kidney biopsy samples on a similar level as pathologists, according to two new studies. https://www.deccanchronicle.com/lifestyle/health-and-wellbeing/080919/artificial-intelligence-augments-traditional-method-of-diagnosing-kidn.html
58 A new ML model from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) can estimate a person's risk of cardiovascular death based on only a raw electrocardiogram (ECG) signal http://news.mit.edu/2019/using-machine-learning-estimate-risk-cardiovascular-death-0912
59 Google developed a deep learning system that can spot common dermatological disorders. https://venturebeat.com/2019/09/13/googles-ai-detects-26-skin-conditions-as-accurately-as-dermatologists/
60 The startup MoooFarm has developed a facial recognition algorithm for cows.  https://www.livemint.com/technology/tech-news/this-company-uses-machine-learning-to-tell-cows-apart-from-each-other-1568386227724.html
61 OpenAI's bots taught themselves skills, such as cooperation and exploiting glitches, to win at hide-and-seek.  https://www.newscientist.com/article/2216410-ai-learns-to-defy-the-laws-of-physics-to-win-at-hide-and-seek/
62 Schlumberger, Chevron, and Microsoft launched an AI platform that aims to speed up projects in the oil industry. https://www.chron.com/business/energy/article/Schlumberger-Chevron-and-Microsoft-launch-14445539.php
63 Pure Storage launched the AI Data Hub, which is co-developed with Nvidia, to "integrate ingestion and inference via Kubernetes orchestration." https://www.zdnet.com/article/pure-storage-launches-ai-data-hub-builds-out-ai-workflow/
64 Researchers developed a new technique using GANs to anonymize people in photos and videos, potentially granting them privacy. https://syncedreview.com/2019/09/16/new-ai-face-anonymization-model-protects-privacy/
65 Facebook on Wednesday unveiled three new models of its Portal internet-connected cameras, which use better AI to track people's movements without sending data back to the company. https://www.wired.com/story/the-facebook-portal-smart-speaker-is-back-now-with-more-ai/
66 Allen Institute for AI (AI2) researchers developed an NLP system that cross-referenced interactions between dietary supplements and other drugs, allowing them to create a searchable database of the information.  https://techcrunch.com/2019/09/19/is-that-supplement-safe-to-take-this-ai-tool-scours-research-to-find-out/?renderMode=ie11
67 A team led by geophysicist Paul Johnson developed an algorithm that they say can predict the start of so-called "slow slip" earthquakes, which occur deeper underground than conventional faster earthquakes and can last weeks at a time.  https://www.quantamagazine.org/artificial-intelligence-takes-on-earthquake-prediction-20190919/
68 Facebook researchers developed an AI system, Fashion++, that uses a deep image-generation neural network to recommend changes to people's outfits, according to a new paper.  https://venturebeat.com/2019/09/18/fashion-uses-ai-to-make-unfashionable-outfits-stylish-with-minimal-tweaks/
69 Grammarly used AI to develop a feature that detects tone in a person's writing.  https://venturebeat.com/2019/09/24/grammarly-uses-ai-to-detect-the-tone-and-tenor-of-your-writing/
70 AICure developed an app that analyzes a person's speech and facial expressions to predict if they have schizophrenia. https://www.newscientist.com/article/2217018-ai-is-learning-to-diagnose-schizophrenia-from-a-smartphone-video/
71 Collaborators created a dance routine that combines choreography and machine learning https://mashable.com/video/ai-dance-routine-real-imaginary/