Search Results for author: Parsa Bagherzadeh

Found 13 papers, 1 papers with code

CLaC at SMM4H 2020: Birth Defect Mention Detection

no code implementations SMM4H (COLING) 2020 Parsa Bagherzadeh, Sabine Bergler

For the detection of personal tweets, where a parent speaks of a child’s birth defect, CLaC combines ELMo word embeddings and gazetteer lists from external resources with a GCNN (for encoding dependencies), in a multi layer, transformer inspired architecture.

Word Embeddings

Multi-input Recurrent Independent Mechanisms for leveraging knowledge sources: Case studies on sentiment analysis and health text mining

no code implementations NAACL (DeeLIO) 2021 Parsa Bagherzadeh, Sabine Bergler

This paper presents a way to inject and leverage existing knowledge from external sources in a Deep Learning environment, extending the recently proposed Recurrent Independent Mechnisms (RIMs) architecture, which comprises a set of interacting yet independent modules.

Sentiment Analysis

Competing Independent Modules for Knowledge Integration and Optimization

no code implementations Findings (EMNLP) 2021 Parsa Bagherzadeh, Sabine Bergler

This paper presents a neural framework of untied independent modules, used here for integrating off the shelf knowledge sources such as language models, lexica, POS information, and dependency relations.

POS Sentiment Analysis

Leveraging knowledge sources for detecting self-reports of particular health issues on social media

no code implementations EACL (Louhi) 2021 Parsa Bagherzadeh, Sabine Bergler

This paper investigates incorporating quality knowledge sources developed by experts for the medical domain as well as syntactic information for classification of tweets into four different health oriented categories.

Classification

CLaC-BP at SemEval-2021 Task 8: SciBERT Plus Rules for MeasEval

no code implementations SEMEVAL 2021 Benjamin Therien, Parsa Bagherzadeh, Sabine Bergler

We analyze ablation experiments and demonstrate how the system components, namely tokenizer, unit identifier, modifier classifier, and language model, affect the overall score.

Language Modelling

CLaC at SemEval-2020 Task 5: Muli-task Stacked Bi-LSTMs

no code implementations SEMEVAL 2020 MinGyou Sung, Parsa Bagherzadeh, Sabine Bergler

We consider detection of the span of antecedents and consequents in argumentative prose a structural, grammatical task.

Multi-Task Learning POS +1

CLaC at SMM4H Task 1, 2, and 4

no code implementations WS 2018 Parsa Bagherzadeh, Nadia Sheikh, Sabine Bergler

CLaC Labs participated in Tasks 1, 2, and 4 using the same base architecture for all tasks with various parameter variations.

Outlier absorbing based on a Bayesian approach

no code implementations2 Jul 2016 Parsa Bagherzadeh, Hadi Sadoghi Yazdi

The presence of outliers is prevalent in machine learning applications and may produce misleading results.

BIG-bench Machine Learning

Cannot find the paper you are looking for? You can Submit a new open access paper.