Deep Learning for Text Analysis Where do we stand?

Slides:



Advertisements
Similar presentations
Talking about origins.
Advertisements

“…by 2014, about 34% of all new business software purchases will be consumed via SaaS…” - IDC, June 2010* Used by Over 50% of the Fortune % CIOs.
Colegio de Bachilleres del Estado de Hidalgo Plantel Tianguistengo Countries and Nationalities By LELI Susana del Carmen Díaz Cañas.
Countries and Nationalities.
NATIONALITIES. Is dim sum Polish? No, it isn’t. It is Chinese.
Revision of unit 1 8th grade. What is the nationality of the USA?
Languages Categorilla
History of Spanish by Rachel and Valentina
Country, Nationality, and Language!. I’m Columbian.
Look at the following expressions and tell us which one you think is better. Then read the introduction to formal and informal English on Page 26.
WORD POWER – UNIT 2 Formal and Informal English. Today’s Agenda  Study the differences of Formal and Informal English.  Practice using Formal and Informal.
Creating With Code.
Developed for English I Inside Out by Donna Rowan.
Countries and nationalities (Europe) Markéta Zakouřilová VY_32_INOVACE_104 ZŠ Jenišovice.
2A. She’s British. Countries & nationalities Point out what countries through their flags & which nationalities?
Países y nacionalidades en inglés
Los países hispanohablantes The Spanish-speaking countries.
South American Countries Review
COUNTRIES AND NATIONALITIES
Capitalization DAY 1 COMPLETE SLIDES 1-8
Visibility and accessibility Accessibility SciELO: Scientific Electronic Library Online ( Financed by the State of Sao Paulo Science Agency.
Countries of the world.
Miscellaneous Freeway - Vol. 2 Unit 5 Varieties of English.
Mark Cieliebak Jan Deriu Dominik Egger Fatih Uzdilli
RECENT TRENDS IN SMT By M.Balamurugan, Phd Research Scholar,
V E R B O T He is… You are … She is… I am… We are…
CS 4501: Introduction to Computer Vision Computer Vision + Natural Language Connelly Barnes Some slides from Fei-Fei Li / Andrej Karpathy / Justin Johnson.
Deep Learning Amin Sobhani.
V E R B O T He is… You are … She is… I am… We are…
COUNTRIES AND FLAGS.
A Hierarchical Model of Reviews for Aspect-based Sentiment Analysis
COUNTRIES AND NATIONALITIES
Giuseppe Attardi Dipartimento di Informatica Università di Pisa
Deep learning and applications to Natural language processing
Travel & Around the World
Deep Learning Workshop
Unsupervised Learning of Tag-Sensitive Descriptors
UNIT 1 Lesson B.
convolutional neural networkS
COUNTRIES; NATIONALITIES; LANGUAGE
Countries and Nationalities
Project # 5 Generating Privacy and Security Threat Summary for Internet of Things REU student: Domonique Cox Graduate mentors: Kaiqiang Song Faculty mentor(s):
Countries nationalities
Giuseppe Attardi Dipartimento di Informatica Università di Pisa

Seminar Topics and Projects
SVM-based Deep Stacking Networks
Countries – Nationalities
Socialized Word Embeddings
Countries – Nationalities
His /her name is ... He / she is from....
Part of Speech Tagging with Neural Architecture Search
Animal adaptations BAAM!.
Culture of Venezuela Najee Murray.
COUNTRİES & NATİONALİTİES
COUNTRIES NATIONALITIES LANGUAGES.
NATIONALITIES  « What’s your nationality? » « I’m French »
NATIONALITIES LANGUAGES
Giuseppe Attardi Dipartimento di Informatica Università di Pisa
Countries and nationalities
Heterogeneous convolutional neural networks for visual recognition
Countries - Nationalities
Idiap Research Institute University of Edinburgh
U-Net: Convolutional Network for Segmentation
COUNTRIES AND NATIONALITIES.

Huawei CBG AI Challenges
BREKFEAST IN WORLD TEXT SI IMAGINI DE PE INTERNET
ARTIFICIAL EUIPO Miguel Ortega Geneva, 27 September 2019.
Presentation transcript:

Deep Learning for Text Analysis Where do we stand? Good morning. My name is Jan and I’m working as a research assistant at the Zurich university of applied sciences in Marks group. Today I will give you an overview of Deep Learning for Text Analysis is used, how it performs and where the challenges and limitations lie. Jan Deriu SwissText Conference, 9th June 2016

Intro

Language Model Illustration: http://sebastianruder.com/word-embeddings-1/

Properties Image credits: https://www.tensorflow.org/tutorials/word2vec

Deep Learning – Convolutional Neural Networks Illustration: Kim, Y. (2014). Convolutional Neural Networks for Sentence Classification

Task 1: Sentiment Analysis - Multilingual

3 Phase Learning Illustration: Deriu, Jan, et al. "Leveraging Large Amounts of Weakly Supervised Data for Multi-Language Sentiment Classification."

Data – Distant Phase

Data – Supervised Phase – SemEval 2016

Results Method English French German Italian SL-CNN 63.49 64.79 65.09 67.79 SL-CNN (no dist.) 60.46 63.25 62.10 64.08 SVM 60.61 - RF 48.60 53.86 52.40 52.71

Competition Winner SemEval 2016 EvalIta 2016

Summary Sentiment Analysis Easy to adapt for multiple languages Data-intensive

Task 2: Gender, Age and Variety

Task 2: Gender and Variety

Data - PAN 2017

Data: Variety Language English Australian Canadian British Irish New Zealand USA Spanish Argentina Chile Colombia Mexico Peru Spain Venezuela Portuguese Brazil Portugal Arabic Egypt Gulf Maghrebi Levantine

Data – Age PAN 2016

F1 scores: GRU – PAN 2017

Results: Architectures (English only)

Results: Architectures (English only)

Results – PAN 2016 (English only)

Summary Good data yields good results In Deep Learning the focus lies in finding and tuning the correct architecture

Task 3: Community Question Answering (cQA)

cQA - Data - SemEval 2017

cQA - Approach - Siamese CNN

cQA – Results SemEval 2017

cQA - Summary Deep Learning supports a large variety of architectures Feautre-based approach works well

Conclusion Deep Learning – very data-intensive Not always better than feature-based approaches From feature-engeneering to «archtiecture»-engeneering

https://github.com/spinningbytes/deep-mlsa Code https://github.com/spinningbytes/deep-mlsa