Context-aware API for natural language understanding and generation - BC-353
Preferred Disciplines: Computer Science, Natural Language Processing, 2nd-year M.Sc. or Ph.D.
Project length: 4-12 months (1 to 2 units)
Desired start date: As soon as possible.
Location: Vancouver, BC
No. of Positions: 1
Preferences: From universities who have an active machine learning / NLP group (e.g. UBC, SFU, etc). Language: English
The partner is a technology startup that intends to improve the process of application development in the area of NLU and conversational agents.
Research and development of an API for large scale enterprise applications in information extraction, intent classification, word sense disambiguation, and question answering dialog systems. The API is a collection of tasks common in natural language processing and automatic text generation.
Background and required skills
- Identifying the proper approach in context modeling (ontology-based, logic-based, CLSTM, etc.)
- Connection of the concepts/entities mentioned in utterances to concepts/facts in reference knowledge bases
- Generation of statements that are correct and relevant within the domain of application, both syntactically and semantically
- To be determined
Expertise and Skills Needed:
- Strong understanding of statistical modeling/inference and machine learning, with focus on natural language processing
- Current familiarity with the academic literature and the state of research
- Advanced knowledge of Python and experience with NLP toolkits such as NLTK, spaCy, gensim, or scikit-learn
- General software abilities such as source control (git), development and test under unix-like environments such as Linux, and working with SQL and NoSQL databases
- Plus but not required: experience of NLU pipeline development, distributed computing and cloud deployment
For more info or to apply to this applied research position, please