About
* PhD in Computer Science (AI/NLU)
* Natural Language Understanding (NLU)
*…
Articles by Walid
-
Why Ambiguity is Necessary, and why Natural Language is not Learnable
Why Ambiguity is Necessary, and why Natural Language is not Learnable
By Walid Saba
-
4 Technical Reasons why Data-Driven and Machine Learning NLU is a Myth
4 Technical Reasons why Data-Driven and Machine Learning NLU is a Myth
By Walid Saba
Contributions
-
What are the most promising NLP applications for text summarization?
Electronic Health Records (EHRs) can be overwhelming - an EHR for a patient over a few years contains an enormous number of notes taken by nurses, physicians, lab technicians, etc. Allowing a physician to look at an intelligent summary of the entire history of a patient can be of great value saving physicians tremendous amount of time.
-
What are the best practices and tips for teaching and learning NLP?
At least for a graduate level NLP course I suggest teaching NLP as a Computational Linguistics course where you allow students to do some digging into various aspects of the field, some of which are not very popular now. While NLP is now dominated by empirical and quantitative (statistical and machine learning) methods, in my opinion students should have some exposure to linguistics, psycholinguistics, formal ontology, the philosophy of language, logical semantics and some cognitive linguistics (and cognitive science issues related to language). Students should have the foundations necessary to be able to pivot if (when?) the dominant paradigm in NLP research does change in more linguistic, ontological, semantic and cognitive directions.
Activity
-
Deskilling machines that can only predict the past very well. That's what they always were and continue to be. And the present or the future? Well…
Deskilling machines that can only predict the past very well. That's what they always were and continue to be. And the present or the future? Well…
Liked by Walid Saba
-
Mozilla recently recognized our Director of Responsible AI Practice Cansu Canca, Ph.D. as one of this year’s Rise25 honorees, highlighting her…
Mozilla recently recognized our Director of Responsible AI Practice Cansu Canca, Ph.D. as one of this year’s Rise25 honorees, highlighting her…
Liked by Walid Saba
Experience
Education
Publications
-
Language and its Commonsense: Where Formal Semantics Went Wrong, and Where it Can (and Should) Go
Journal of Knowledge Structures and Systems (JKSS)
The purpose of this paper is twofold: (i) we will argue that formal semantics
might have faltered due to its failure in distinguishing between two fundamentally very different types of concepts, namely ontological concepts, that should
be types in a strongly-typed ontology, and logical concepts, that are predicates
corresponding to properties of, and relations between, objects of various ontological types; and (ii) we show that accounting for these differences amounts to
a new…The purpose of this paper is twofold: (i) we will argue that formal semantics
might have faltered due to its failure in distinguishing between two fundamentally very different types of concepts, namely ontological concepts, that should
be types in a strongly-typed ontology, and logical concepts, that are predicates
corresponding to properties of, and relations between, objects of various ontological types; and (ii) we show that accounting for these differences amounts to
a new formal semantics; one that integrates lexical and compositional semantics
in one coherent framework and one where formal semantics is embedded with a
strongly typed ontology; an ontology that reflects our commonsense knowledge
of the world and the way we talk about it in ordinary language. We will show
how in such a framework a number of challenges in the semantics of natural
language are adequately and systematically treated. -
On the Winograd Schema: Situating Language Understanding in the Data-Information-Knowledge Continuum
AAAI Press
The Winograd Schema (WS) challenge, proposed as an al-ternative to the Turing Test, has become the new standard for evaluating progress in natural language understanding (NLU). In this paper we will not however be concerned with how this challenge might be addressed. Instead, our aim here is threefold: (i) we will first formally 'situate' the WS challenge in the data-information-knowledge continuum, suggesting where in that continuum a good WS resides; (ii) we will show that a WS is just…
The Winograd Schema (WS) challenge, proposed as an al-ternative to the Turing Test, has become the new standard for evaluating progress in natural language understanding (NLU). In this paper we will not however be concerned with how this challenge might be addressed. Instead, our aim here is threefold: (i) we will first formally 'situate' the WS challenge in the data-information-knowledge continuum, suggesting where in that continuum a good WS resides; (ii) we will show that a WS is just special case of a more general phenomenon in language understanding, namely the missing text phenomenon (henceforth, MTP) - in particular, we will argue that what we usually call thinking in the process of language understanding involves discovering a significant amount of 'missing text' - text that is not explicitly stated, but is often implicitly assumed as shared background knowledge; and (iii) we conclude by a brief discussion on why MTP is inconsistent with the data-driven and machine learning approach to language understanding.
-
Logical Semantics and Commonsense Knowledge: Where Did we Go Wrong, and How to Go Forward, Again
arXiv.org
The purpose of this paper is twofold: (i) we will argue that logical semantics might have faltered due to its failure in distinguishing between two fundamentally very different types of concepts, namely ontological concepts, that should be types in a strongly-typed ontology, and logical concepts, that are predicates corresponding to properties of and relations between objects of various ontological types; and (ii) we will show that accounting for these differences amounts to the integration of…
The purpose of this paper is twofold: (i) we will argue that logical semantics might have faltered due to its failure in distinguishing between two fundamentally very different types of concepts, namely ontological concepts, that should be types in a strongly-typed ontology, and logical concepts, that are predicates corresponding to properties of and relations between objects of various ontological types; and (ii) we will show that accounting for these differences amounts to the integration of lexical and compositional semantics in one coherent framework, and to an embedding in our logical semantics of a strongly-typed ontology, an ontology that reflects our commonsense view of the world and the way we talk about it in ordinary language. In such a framework, as will be demonstrated, a number of challenges in the semantics of natural language can be ade-quately and systematically treated.
-
Commonsense Knowledge, Ontology and Ordinary Language
Int. Journal of Reasoning-based Intelligent Systems (IJRIS), 2 (1), pp. 36-50.
Over two decades ago, a 'quiet revolution' overwhelmingly replaced knowledge-based approaches in natural language processing (NLP) by quantitative (e.g., statistical, corpus-based, machine learning) methods. Although it is our firm belief that purely quantitative approaches cannot be the only paradigm for NLP, dissatisfaction with purely engineering approaches to the construction of large knowledge bases for NLP are somewhat justified. In this paper we hope to demonstrate that both trends are…
Over two decades ago, a 'quiet revolution' overwhelmingly replaced knowledge-based approaches in natural language processing (NLP) by quantitative (e.g., statistical, corpus-based, machine learning) methods. Although it is our firm belief that purely quantitative approaches cannot be the only paradigm for NLP, dissatisfaction with purely engineering approaches to the construction of large knowledge bases for NLP are somewhat justified. In this paper we hope to demonstrate that both trends are partly misguided and that the time has come to enrich logical semantics with an ontological structure that reflects our commonsense view of the world and the way we talk about in ordinary language.
-
Polymorphism as a model for ambiguity in natural language
KI – the German Journal of Artificial Intelligence
We suggest modelling concepts as types in a strongly-typed ontology that reflects our commonsense view of the world and the way we talk about in ordinary spoken language. In such a framework, certain types of ambiguities in natural language are explained by the notion of polymoprhism.
-
Language, Logic & Ontology: Uncovering the Structure of Commonsense Knowledge
International Journal of Human-Computer Studies, 65 (7), pp. 610-623
The purpose of this paper is twofold: (i) we argue that the structure of commonsense knowledge must be discovered, rather than invented; and (ii) we argue that natural language, which is the best known theory of our (shared) commonsense knowledge, should itself be used as a guide to discovering the structure of commonsense knowledge. In addition to suggesting a systematic method to the discovery of the structure of commonsense knowledge, the method we propose seems to also provide an…
The purpose of this paper is twofold: (i) we argue that the structure of commonsense knowledge must be discovered, rather than invented; and (ii) we argue that natural language, which is the best known theory of our (shared) commonsense knowledge, should itself be used as a guide to discovering the structure of commonsense knowledge. In addition to suggesting a systematic method to the discovery of the structure of commonsense knowledge, the method we propose seems to also provide an explanation for a number of phenomena in natural language, such as metaphor, intensionality, and the semantics of nominal compounds. Admittedly, our ultimate goal is quite ambitious, and it is no less than the systematic ‘discovery’ of a well-typed ontology of commonsense knowledge, and the subsequent formulation of the long-awaited goal of a meaning algebra.
-
Plausible Reasoning and the Resolution of Quantifier Scope Ambiguities
Studia Logica, March 2001, Volume 67, Issue 2, pp 271-289
Despite overwhelming evidence suggesting that quantifier scope is a phenomenon that must be treated at the pragmatic level, most computational treatments of scope ambiguities have thus far been a collection of syntactically motivated preference rules. This might be in part due to the prevailing wisdom that a commonsense inferencing strategy would require the storage of and reasoning with a vast amount of background knowledge. In this paper we hope to demonstrate that the challenge in developing…
Despite overwhelming evidence suggesting that quantifier scope is a phenomenon that must be treated at the pragmatic level, most computational treatments of scope ambiguities have thus far been a collection of syntactically motivated preference rules. This might be in part due to the prevailing wisdom that a commonsense inferencing strategy would require the storage of and reasoning with a vast amount of background knowledge. In this paper we hope to demonstrate that the challenge in developing a commonsense inferencing strategy is in the discovery of the relevant commonsense data and in a proper formulation of the inferencing strategy itself, and that a massive amount of background knowledge is not always required. In particular, we present a very effective procedure for resolving quantifier scope ambiguities at the pragmatic level using simple ‘quantitative’ data that is readily available in most database environments.
-
Modeling Mental States in Agent Negotiation
AAAI - Compilation copyright © 2001, AAAI (www.aaai.org). All rights reserved
Patents
-
Techniques for understanding the aboutness of text based on semantic analysis
Issued US US20160292145 A1
In one embodiment of the present invention, a semantic analyzer translates a text segment into a structured representation that conveys the meaning of the text segment. Notably, the semantic analyzer leverages a semantic network to perform word sense disambiguation operations that map text words included in the text segment into concepts—word senses with a single, specific meaning—that are interconnected with relevance ratings. A topic generator then creates topics on-the-fly that includes one…
In one embodiment of the present invention, a semantic analyzer translates a text segment into a structured representation that conveys the meaning of the text segment. Notably, the semantic analyzer leverages a semantic network to perform word sense disambiguation operations that map text words included in the text segment into concepts—word senses with a single, specific meaning—that are interconnected with relevance ratings. A topic generator then creates topics on-the-fly that includes one or more mapped concepts that are related within the context of the text segment. In this fashion, the topic generator tailors the semantic network to the text segment. A topic analyzer processes this tailored semantic network, generating a relevance-ranked list of topics as a meaningful proxy for the text segment. Advantageously, operating at the level of concepts and topics reduces the misinterpretations attributable to key word and statistical analysis methods.
Other inventorsSee patent
More activity by Walid
-
The irony with artificial neural networks/machine learning, now called AI, from the start has been that these models were advertised for making…
The irony with artificial neural networks/machine learning, now called AI, from the start has been that these models were advertised for making…
Liked by Walid Saba
-
Ready to take your health tech startup to the next level? Apply now to join the next cohort of the Future of Healthcare Founder Residency. Designed…
Ready to take your health tech startup to the next level? Apply now to join the next cohort of the Future of Healthcare Founder Residency. Designed…
Liked by Walid Saba
-
🌟🌟🌟 I'm thrilled to announce that our paper, "Analyzing Cultural Representations of Emotions in LLMs through Mixed Emotion Survey," has been…
🌟🌟🌟 I'm thrilled to announce that our paper, "Analyzing Cultural Representations of Emotions in LLMs through Mixed Emotion Survey," has been…
Liked by Walid Saba
-
I am proud of this work. Thank you for leading it, Shiran Dudy.
I am proud of this work. Thank you for leading it, Shiran Dudy.
Liked by Walid Saba
-
Few Philosophy Bears Got Laid Off = Australian Catholic University (ACU) has finalized its “Academic Change Management Plan” and the closure of the…
Few Philosophy Bears Got Laid Off = Australian Catholic University (ACU) has finalized its “Academic Change Management Plan” and the closure of the…
Liked by Walid Saba
-
Last week I co-chaired ACM KDD 2024 with Francesco Bonchi with more than 2,300 attendees, a complete success. The main highlights for me were: (1)…
Last week I co-chaired ACM KDD 2024 with Francesco Bonchi with more than 2,300 attendees, a complete success. The main highlights for me were: (1)…
Liked by Walid Saba
-
In the latest episode of the podcast "Legends of Data & AI", host Usama Fayyad from the Institute for Experiential AI at Northeastern University…
In the latest episode of the podcast "Legends of Data & AI", host Usama Fayyad from the Institute for Experiential AI at Northeastern University…
Liked by Walid Saba
-
New York Magazine presents an unflattering description of the AI influencer #culture, telling funny stories about OpenAI’s imminent form of AGI…
New York Magazine presents an unflattering description of the AI influencer #culture, telling funny stories about OpenAI’s imminent form of AGI…
Liked by Walid Saba
-
HAPPENING TOMORROW: Our Director of Research Ricardo Baeza-Yates will deliver a keynote at #ASONAM2024 about the limitations of supervised…
HAPPENING TOMORROW: Our Director of Research Ricardo Baeza-Yates will deliver a keynote at #ASONAM2024 about the limitations of supervised…
Liked by Walid Saba
-
🎉 Felicitaciones a Ricardo Baeza-Yates, Premio Nacional de Ciencias Aplicadas y Tecnológicas 2024 🎉 Desde CHILETEC - Asociación de Empresas…
🎉 Felicitaciones a Ricardo Baeza-Yates, Premio Nacional de Ciencias Aplicadas y Tecnológicas 2024 🎉 Desde CHILETEC - Asociación de Empresas…
Liked by Walid Saba
-
This week, I enjoyed attending KDD’24, one of the leading conferences in data mining, data science, and machine learning. This year was particularly…
This week, I enjoyed attending KDD’24, one of the leading conferences in data mining, data science, and machine learning. This year was particularly…
Liked by Walid Saba
Other similar profiles
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More