Kfir Bar

Chief Scientist, Basis Technology

I am the Chief Scientist at Basis Technology. I have spent many years working in a wide range of natural language processing (NLP) disciplines, including statistical machine translation, named entity recognition, and digital-humanity applications. I am a big fan of combining linguistic knowledge with sophisticated AI algorithms, for extracting the most important information from a piece of text. Before Basis, I worked for Intuview as CTO, supporting national security and counter-terrorism missions by deducing authorship, sentiment, intent, and other contextual information. In 2013, I co-founded Comprendi, which transforms big data into actionable marketing insights. We served some large scale advertisers in different verticals, with great success. At 2016 our team won the 2016 Twitter #Promote challenge, for a $250K prize in cash. I am a lecturer at Tel Aviv University, COMAS and IDC, where I teach courses in computer science, digital humanities, machine learning and natural language processing. I hold a PhD in computer science from Tel Aviv University for a thesis on Semantics and Machine Translation, titled "Deriving Paraphrases for Highly Inflected Languages, with a Focus on Machine Translation .

Videos:

My talk at HLTCon, Washington DC, Nov 29, 2018.

My talk at #ODSC Europe, London, Sep 22, 2018.

Latest updates:

My talk at Money20/20 Asia, Singapore, March 19, 2019.

Speaking at Suits and Spooks, Washington DC, Jan 25, 2019.

I spoke at #ODSC Europe, London, Sep 22, 2018. Here are my Slides

Speaking at the Diffeo Analytical Tradecraft and Technology Conference, Washington DC, June 6, 2018.

Paper accepted (CLPsych @NAACL 2019): Semantic Characteristics of Schizophrenic Speech , Kfir Bar, Vered Zilberstein, Ido Ziv, Heli Baram, Nachum Dershowitz, Samuel Itzikowitz, Eiran Vadim Harel, June, 2019

Education.

Activities.

June, 2018

Diffeo Analytical Tradecraft and Technology Forum

My talk about connecting names with deep learning, at the Diffeo Analytical Tradecraft and Technology Forum

Connecting Names with Deep Learning

A cornerstone of customer relationship management, chatbot analytics, and research automation systems, Named Entity Recognition (NER) is a key commercial application of Natural Language Processing (NLP). State of the art approaches to NER are purely data driven, leveraging deep neural networks to identify named entity mentions—such as people, organizations, and locations—in lakes of text data. In this talk, I will present the latest academic research on NER and provide real-life examples of how we are applying these cutting-edge techniques to multiple languages, including English, Arabic, Korean, and Japanese. We'll look at accuracy, speed, and memory footprint, while comparing some of the best known deep architectures with a basic statistical approach.

Relevant links:

Diffeo Analytical Tradecraft and Technology Forum

April, 2018

AI and NLP: Data Talks

My talk about Deep Learning for Named Entity Recognition at the Tokyo Artificial Intelligence and NLP Society

Deep Learning for Named Entity Recognition

Named Entity Recognition is one of the key tasks in commercial Natural Language Processing applications. Its objective is to identify named entity mentions, such as people, organizations, and locations, in running text. State-of-the-art approaches are purely data-driven, leveraging deep neural networks. In this talk, I will present a few of those works, followed by a description of our own deep NER implementation. We'll look at accuracy, speed, and memory footprint, while comparing some of the best known deep architectures with a basic statistical approach.

Relevant links:

AI and NLP: Data Talks Tokyo Artificial Intelligence and NLP Society

April, 2016

The 2016 Twitter #Promote challenge

Comprendi Wins $250,000 Grand Prize in Twitter Innovation Competition

Comprendi Wins $250,000 Grand Prize in Twitter Innovation Competition>

Comprendi is the grand prize winner of the Twitter #Promote Ads API Challenge for its groundbreaking advertising automation solution driven by artificial intelligence and big data. At an awards ceremony held at Twitter headquarters in San Francisco, Comprendi was recognized for its new Adaptive Creative technology powered by the Twitter Ads API and Twitter’s unique Firehose data API.

Relevant links:

#Promote Ads API Challenge 2016: Announcing the Regional Finalists and Grand Prize Winner

June, 2015

Active Learning for NLP

My talk about Active Learning for NLP at The Israeli Natural Language Processing Meetup

As a co-founder and CTO at Comprendi, I gave a talk about our experience with the subject of Active Learning for NLP

Relevant links:

Active Learning for NLP The Israeli Natural Language Processing Meetup

Comprendi

I am interested in Natural Language Processing, Deep Learning, Machine Learning, Digital Humanitieis, and Computational Creativity.

Publications.

Metaphor Interpretation Using Word Embeddings

CICLING, Hanoi, Vietnam, 2018

Kfir Bar, Nachume Dershowitz, Lena Dankin

Will appear soon

SLS at SemEval-2016 Task 3: Neural-based Approaches for Ranking in Community Question Answering

In Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), San Diego, 2016

Mitra Mohtarami, Yonatan Belinkov, Wei-Ning Hsu, Yu Zhang Tao Lei, Kfir Bar, Scott Cyphers, James Glass

Read the article

Processing Judeo-Arabic Texts

In Proceedings of the First Arabic Natural Language Processing Conference, IEEE, Cairo, 2015

Kfir Bar, Nachum Dershowitz, Lior Wolf, Yackov Lubarsky, and Yaacov Choueka

Read the article

The Tel Aviv University System for the Code-Switching Workshop Shared Task

In Proceedings of the First Workshop on Computational Approaches to Code-Switching, Conference on Empirical Methods on Natural Language Processing (EMNLP), Doha, Qatar, 2014

Kfir Bar and Nachum Dershowitz

Read the article

Inferring Paraphrases for a Highly Inflected Language from a Monolingual Corpus

In Proceedings of the Conference on Intelligent Text Processing and Computational Linguistics (CICLing), Part 2, Lecture Notes in Computer Science, vol. 8404, Springer-Verlag, Heidelberg, pp. 254-270, Kathmandu, Nepal, 2014

Kfir Bar and Nachum Dershowitz

Read the article

Joint word2vec Networks for Bilingual Semantic Representations

In Proceedings of the Conference on Intelligent Text Processing and Computational Linguistics (CICLing), Kathmandu, Nepal, 2014

Lior Wolf, Yair Hanani, Kfir Bar, and Nachum Dershowitz

Read the article

Deriving Paraphrases for Highly-Inflected Languages from Comparable Documents

In Proceedings of the 24th International Conference on Computational Linguistics (COLING), Mumbai, India, 2012

Kfir Bar and Nachum Dershowitz

Read the article

Building an Arabic Multiword Expressions Repository

In Proceedings of the ACL 2012 Joint Workshop on Statistical Parsing and Semantic Processing of Morphologically Rich Languages, Jeju, Korea, 2012

Abdelati Hawwari, Kfir Bar and Mona Diab

Read the article

Using Synonyms for Arabic-to-English Example-Based Translation

In Proceedings of The Ninth Conference of the Association for Machine Translation in the Americas (AMTA-9), Denver, CO, USA, 2010

Kfir Bar and Nachum Dershowitz

Read the article

Automatically Classifying Documents by Ideological and Organizational Affiliation

In Proceedings of the 2009 IEEE international conference on Intelligence and security informatics (ISI'09), Arizona, USA, 2009

Moshe Koppel, Navot Akiva, Eli Alshech and Kfir Bar

Read the article

Example-Based Arabic-to-English Translation

In Proceedings of a Workshop on Arabic Natural Language Processing, pp. 325/1-4, Fez, Morocco, 2007

Kfir Bar, Nachum Dershowitz, and Yaacov Choueka

Read the article

Teaching.

Contact.

barkfir at yahoo dot com skype: kfirbar_skype

Kfir Bar. Copyright 2018.