CSCI-B 659 Topics in Artificial Intelligence

LING-L 715 Semantic Natural Language Processing, NoAI, and Big Knowledge (including dialog and AI)

This is the course page for Topics in Artificial Intelligence / Semantic Natural Language Processing, NoAI, and Big Knowledge (including dialog and AI) by Damir Cavar.

Damir Cavar

– August 2020 –

Introduction

You might wonder, why does the title mention NoAI and Big Knowledge. The reason for NoAI is that the use of the term AI nowadays has drifted away from its core meaning, that we have to call our approach to artificial intelligence NoAI. The term Big Knowledge as opposed to Big Data should emphasize the fact that data as such is less relevant, at least for us here. What we are interested in is aggregation of information and the validation, transformation, conversion of information to knowledge. Big Data was a necessary step that involved a significant computational challenge that consequently leads us towards the real goal and challenge: Big Knowledge and all the related theoretical and computational challenges.

The goal of this seminar is to understand the limitations of NLP, Big Knowledge, NoAI (and the modern form of AI), to establish a research and education platform, and to design a research agenda for the next years.

Support

This project is supported by Amazon AWS Educate with AWS Promotional Credits to the professor and various students.

State

We have set up an Amazon AWS EC2 server instance at the URL:

http://linguistic.technology

The server will provide more detailed information about the technologies and the current status of development. The general architecture involves:

NLP pipelines for parsing text and linguistic annotation of the content at all levels, including sentence splitting, tokenization, part of speech tagging, lemmatisation, syntactic parsing (constituent structure and phrase structure), dependency parsing, anaphora resolution, co-reference analysis, semantic concept mapping, and so on.
Mapping of extracted concepts and relations to knowledge graphs using Graph-DBs, with OWL and reasoner supported semantic technologies.
Graph-DB storage for knowledge graphs and concept networks.
Question-Answering using speech recognition and existing AI infrastructure (Alexa, Cortana, Google Assistant, etc.), and mapping of spoken language queries to SPARQL, mapping of SPARQL-results to natural language output/text using text to speech.

Literature

In general, useful textbooks for beginners in speech and language processing, Computational Linguistics (CL), Natural Language Processing (NLP), or even AI (language related) are:

Chris Manning and Hinrich Schütze (1999) Foundations of Statistical Natural Language Processing. MIT Press, Cambridge, MA. (This is somewhat outdated and has weak points, but it is a good start for absolute beginners.)
Dan Jurafsky and James H. Martin (2017, 3rd edition draft) Speech and Language Processing. (This is my preferred textbook, the 3rd edition draft is available online as PDF.)

A good idea about very concrete topics in CL/NLP you can get from:

Ruslan Mitkov (2nd edition in preparation) The Oxford Handbook of Computational Linguistics. (The 1st edition should be in your library or in your bookshelf.)

If you are interested in NLP, but you have little or no background in linguistics, a very nice and useful introduction is:

Emily M. Bender (2013) Linguistic Fundamentals for Natural Language Processing: 100 Essentials from Morphology and Syntax. Morgan and Claypool Publishers.

You might soon discover that linguistics is an absolutely interesting field and you might want to know more about your language, why you speak and how you do it, and so on. There is a large number of very good linguistic textbooks, courses, general learning material. One strategy is to consult your local linguist. Another is to check out linguistic programs, the courses covered, the course material etc. Here is a random selection of very good linguistic textbooks to start with:

Adrian Akmajian, Richard A. Demers, Ann K. Farmer and Robert M. Harnish (2010) Linguistics (6th edition). MIT Press.
More specific and advanced for selected topics: Oxford Textbooks in Linguistics

There is also an interesting Wikibooks project as part of the Linguistics Collection, which might provide valuable information to start with:

Wikibooks: Linguistics