About
I have 30 years of info-tech industry experience across the entire software development life-cycle. I am an experienced Artificial Intelligence researcher and product manager with polished client-facing skills, who can lead an engineering team to deliver projects within the triple constraints (cost, time, and scope).
My PhD from Cambridge University (UK) is in computational linguistics. I have a decade of experience in information retrieval (search technologies), machine learning, data mining, and analytics.
I am open to consulting opportunities.
Specialties: Intelligent systems development, with special interests in computational linguistics (search technologies, machine translation, summarization)
Experienced with machine learning, in-simulation competency assessment, natural language generation, decision support system, analytics, event identification and data visualization.
Articles by Shahzad
Activity
-
Javed Iqbal successfully defended his thesis titled " Domain Adaptation for Semantic Segmentation." Javed Iqbal, now Dr. Javed Iqbal, is my first…
Javed Iqbal successfully defended his thesis titled " Domain Adaptation for Semantic Segmentation." Javed Iqbal, now Dr. Javed Iqbal, is my first…
Liked by Shahzad Khan, PhD
-
Apology and clarification: I posted earlier today that Noon turned a profit in 2023 based on details shared in the Public Investment Fund's annual…
Apology and clarification: I posted earlier today that Noon turned a profit in 2023 based on details shared in the Public Investment Fund's annual…
Liked by Shahzad Khan, PhD
-
Looking for someone with Product Management experience who wants to work in Business Development. Or someone in Business Development who can think…
Looking for someone with Product Management experience who wants to work in Business Development. Or someone in Business Development who can think…
Liked by Shahzad Khan, PhD
Experience
Education
-
-
My area of research was in sentiment classification using linguistic analyses and machine learning. The title of my research thesis is 'Negation and Antonymy in Sentiment Classification'. I carried out my research as part of the <a href=http://www.cl.cam.ac.uk/research/>Natural Language and Information Processing</a> group under the supervision of Dr. Ted Briscoe.
-
-
Activities and Societies: Phi Beta Delta International Honor Society
The Telecommunications and Network Management graduate degree offered at the School of Information Studies is a marriage of technology, policy and management. This programme gives IT professionals a broader view of how to apply and use networking technologies to solve business problems, provide strategic planning and consultancy, and design customer and business-need driven winning products.
-
-
-
-
Volunteer Experience
-
Member
Member of the Task Force on Applications of AI in Pakistan
- Present 1 year 8 months
Advising on the development of modern R&D capabilities, formation of practical regulations and establishment of commercialization efforts through international collaborations
-
Advisor, CET-CS/CP/WDIA Program Advisory Committee
Algonquin College of Applied Arts and Technology
- Present 3 years 4 months
Helping Algonquin College to keep it's AI Certificate curriculum relevant to industry needs as an advisory members of the curriculum oversight committee
-
Advisor, Computer Science, Electrical Engineering Curriculum Advisory Board
University of Ottawa
- Present 1 year 7 months
Helping the University of Ottawa Computer Science and Electrical Engineering Programs to be relevant to the needs of industry
Publications
-
Relevance feedback using semantic association between indexing terms in large free text corpuses
IADIS International Conference WWW/Internet - ICWI
Relevance feedback has been considered as a means of incorporating learning into information retrieval systems for quite sometime now. This paper discusses the research results of two methodologies for automatic query expansion to incorporate relevance feedback into our probabilistic information retrieval system. The first methodology employs analysis of topmost ranked documents retrieved by the user’s initial query to form new keyword enriched expanded queries. These queries help the user in…
Relevance feedback has been considered as a means of incorporating learning into information retrieval systems for quite sometime now. This paper discusses the research results of two methodologies for automatic query expansion to incorporate relevance feedback into our probabilistic information retrieval system. The first methodology employs analysis of topmost ranked documents retrieved by the user’s initial query to form new keyword enriched expanded queries. These queries help the user in focusing his search to a smaller subset of the document collection. The second novel technique developed during the course of this research, resorts to a process similar to text mining employing ‘index term association’ as an underlying mechanism to expand the query with a semantic structure. We also draw a comparison of the two different methodologies using a human judge, measuring the extent of subject deviation of the expanded query from the initial/original query. The results show that significant improvement can be expected using semantic association between indexing terms in query expansion.
Other authorsSee publication -
Statistical Word Sense Disambiguation through Unsupervised Sense Marker Enrichment on Large Free Text Corpuses
IADIS International Conference WWW/INTERNET
Word sense ambiguity is known to have a destructive effect on the performance of information retrieval and linguistic systems. The problem arises from the inherent polysemous nature of natural languages, where one word can have multiple meanings or senses. This is not a problem for humans but mapping the correct sense of a word is a daunting task for a retrieval system. This paper describes two disambiguation methodologies based on contemporary techniques that seek to enrich text with sense…
Word sense ambiguity is known to have a destructive effect on the performance of information retrieval and linguistic systems. The problem arises from the inherent polysemous nature of natural languages, where one word can have multiple meanings or senses. This is not a problem for humans but mapping the correct sense of a word is a daunting task for a retrieval system. This paper describes two disambiguation methodologies based on contemporary techniques that seek to enrich text with sense meta-information by identifying the correct sense for an ambiguous noun in a document. This research draws on contemporary statistical disambiguation methodologies, and attempts to make it more effective through a novel weighting scheme, which is simpler than complex schemes used by other disambiguation algorithms. This research follows two recent ground breaking research results --- that words tend to have one sense per document and one sense per collocation. In the experiments, the set of senses for each polysemous word are the same as the Wordnet 1.7 repository. However, the methodologies are generalized, and applicable to any concept repository that is built on a generalization/specialization framework. The two different methodologies are compared with each other and the results establish that this approach leads to an improvement in the disambiguation process. This paper also proposes a strategy to use the disambiguation methodology to enhance relevance feedback and information retrieval performance.
Other authorsSee publication
Patents
-
Method and system relating to re-labelling multi-document clusters
Issued CA CA2865184C
Individuals receive overwhelming barrage of information which must be filtered, processed, analysed, reviewed, consolidated and distributed or acted upon. However, prior art tools for automatically processing content, such as for example returning search results from an Internet or database search for example are ineffective. Prior art search techniques merely provide large numbers of "hits" with at most removal of multiple occurrences of identical items. However, it would be beneficial to…
Individuals receive overwhelming barrage of information which must be filtered, processed, analysed, reviewed, consolidated and distributed or acted upon. However, prior art tools for automatically processing content, such as for example returning search results from an Internet or database search for example are ineffective. Prior art search techniques merely provide large numbers of "hits" with at most removal of multiple occurrences of identical items. However, it would be beneficial to present searches as a series of multi-document clusters wherein occurrences of commonly themed content are clustered allowing the user to rapidly see the number of different themes and review a selected theme. Further, it would be beneficial, in repeated searches, for new clusters to be identified automatically as well as new items of content associated with existing clusters to be associated to these clusters.
-
Method and system relating to re-labelling multi-document clusters
Issued US US9600470B2
Individuals receive overwhelming barrage of information which must be filtered, processed, analyzed, reviewed, consolidated and distributed or acted upon. However, prior art tools for automatically processing content, such as for example returning search results from an Internet or database search for example are ineffective. Prior art search techniques merely provide large numbers of “hits” with at most removal of multiple occurrences of identical items. However, it would be beneficial to…
Individuals receive overwhelming barrage of information which must be filtered, processed, analyzed, reviewed, consolidated and distributed or acted upon. However, prior art tools for automatically processing content, such as for example returning search results from an Internet or database search for example are ineffective. Prior art search techniques merely provide large numbers of “hits” with at most removal of multiple occurrences of identical items. However, it would be beneficial to present searches as a series of multi-document clusters wherein occurrences of commonly themed content are clustered allowing the user to rapidly see the number of different themes and review a selected theme. Further, it would be beneficial, in repeated searches, for new clusters to be identified automatically as well as new items of content associated with existing clusters to be associated to these clusters.
-
Method and system relating to salient content extraction for electronic content
Issued US US9336202B2
Individuals receive overwhelming barrage of information which must be filtered, processed, analyzed, reviewed, consolidated and distributed or acted upon. Automatic approaches to “scraping” salient content from sources of content are provided allowing the salient content to be provided to the user or subjected to further processing such as clustering or sentiment analysis for example.
Embodiments of the invention provide for:
automated scraper induction based on document and/or contextual…Individuals receive overwhelming barrage of information which must be filtered, processed, analyzed, reviewed, consolidated and distributed or acted upon. Automatic approaches to “scraping” salient content from sources of content are provided allowing the salient content to be provided to the user or subjected to further processing such as clustering or sentiment analysis for example.
Embodiments of the invention provide for:
automated scraper induction based on document and/or contextual semantic cues and document structure analysis.
identifying salient text, removing boiler-plate text, off-topic content and other non-salient content;
deriving reusable descriptive extraction patterns for subsequent documents;
applying descriptive extraction patterns for extraction from subsequent documents from the same source;
intelligent identification of extraction success confidence score, using historical success scores; and
employing confidence scores to automatically trigger new extraction pattern identification if extracted confidence is below an acceptable confidence threshold. -
Method and system relating to sentiment analysis of electronic content
Issued US CA2865186C
Embodiments of the invention provide automatic contextual based sentiment classification of content in terms of both sentiments expressed and their intensity. Further, a content set is analysed to rapidly establish an "at-a-glance" type assessment of the key topics/themes present within the content set and sentimentally annotate each. Importantly, embodiments of the invention also provide for a user to establish the basis for the sentiment associated with an item of or set of content, i.e. make…
Embodiments of the invention provide automatic contextual based sentiment classification of content in terms of both sentiments expressed and their intensity. Further, a content set is analysed to rapidly establish an "at-a-glance" type assessment of the key topics/themes present within the content set and sentimentally annotate each. Importantly, embodiments of the invention also provide for a user to establish the basis for the sentiment associated with an item of or set of content, i.e. make it explainable. Further embodiments of the invention provide for the establishment of psychological tone to sentiments where the sentiments and psychological tones to be tuned are from the context or domain of the content.
-
Method and system relating to salient content extraction for electronic content
Issued CA CA2865187C
Automatic approaches to scraping salient content from sources of content are provided that allow the salient content to be provided to the user or subjected for further processing such as clustering or sentiment analysis. Embodiments of the invention provide for: automated scraper induction based on document and/or contextual semantic cues and document structure analysis; identifying salient text, removing boiler-plate text, off-topic content and other non-salient content; deriving reusable…
Automatic approaches to scraping salient content from sources of content are provided that allow the salient content to be provided to the user or subjected for further processing such as clustering or sentiment analysis. Embodiments of the invention provide for: automated scraper induction based on document and/or contextual semantic cues and document structure analysis; identifying salient text, removing boiler-plate text, off-topic content and other non-salient content; deriving reusable descriptive extraction patterns for subsequent documents; applying descriptive extraction patterns for extraction from subsequent documents form the same source; intelligent identification of extraction success confidence score, using historical success scores; and employing confidence scores to automatically trigger new extraction pattern identification if extracted confidence is below an acceptable confidence threshold.
Languages
-
Turkish
Native or bilingual proficiency
-
Urdu
Native or bilingual proficiency
-
Pashto
Professional working proficiency
-
English
Native or bilingual proficiency
Recommendations received
-
LinkedIn User
2 people have recommended Shahzad
Join now to viewMore activity by Shahzad
-
A Big Win for Ottawa’: Air Canada resumes non-stop flight to London Heathrow from YOW • Read the Story by Sarah M., Reporter, Ottawa Business…
A Big Win for Ottawa’: Air Canada resumes non-stop flight to London Heathrow from YOW • Read the Story by Sarah M., Reporter, Ottawa Business…
Liked by Shahzad Khan, PhD
-
It was refreshing to reconnect with old friends Dr Les Cottrel, Barbara and Umar kalim at a dinner at Roma, Menlo Park CA. Reconnecting with old…
It was refreshing to reconnect with old friends Dr Les Cottrel, Barbara and Umar kalim at a dinner at Roma, Menlo Park CA. Reconnecting with old…
Liked by Shahzad Khan, PhD
-
“Revenue is not the important thing.” That’s what a founder who built in stealth for a year said. Here's how he grew to 1M paid users & raised $130M:…
“Revenue is not the important thing.” That’s what a founder who built in stealth for a year said. Here's how he grew to 1M paid users & raised $130M:…
Liked by Shahzad Khan, PhD
-
🌍 Investing in a Top-Ranked Passport: A Pathway to Career Development 🌍 In today’s globalized world, obtaining a top-ranked passport is one of the…
🌍 Investing in a Top-Ranked Passport: A Pathway to Career Development 🌍 In today’s globalized world, obtaining a top-ranked passport is one of the…
Liked by Shahzad Khan, PhD
-
About 1 year ago, we made a major pivot at Fellow.app and went all in on AI powered meetings (AI Notetaking, Action-item tracking and more). And it's…
About 1 year ago, we made a major pivot at Fellow.app and went all in on AI powered meetings (AI Notetaking, Action-item tracking and more). And it's…
Liked by Shahzad Khan, PhD
-
Don’t stack post-money SAFEs. I know YC backs them & I, as an investor, love them. But for founders, 1 post-money SAFE is fine—2+ can be…
Don’t stack post-money SAFEs. I know YC backs them & I, as an investor, love them. But for founders, 1 post-money SAFE is fine—2+ can be…
Liked by Shahzad Khan, PhD
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 MoreOthers named Shahzad Khan, PhD
2 others named Shahzad Khan, PhD are on LinkedIn
See others named Shahzad Khan, PhD