Abbas Akkasi

Abbas Akkasi

He/Him

PhD-level AI scientist specializing in Generative AI, NLP, and Applied Machine Learning.

Teaching — Abbas Akkasi

Selected Courses and Topics Taught.

Carleton University — Ottawa, Canada
Fall 2025

Introduction to Deep Learning

Undergraduate / Introductory
  • Introduction to Deep Learning
  • Mathematics for Machine Learning / Deep Learning
  • Basics of Machine Learning
  • Introduction to Neural Networks
  • Deep Neural Networks
  • Optimization Algorithms (SGD, RMSprop, Adam)
  • Hyperparameter Tuning
  • Convolutional Neural Networks (CNNs)
  • Recurrent Neural Networks (RNNs)
  • LSTM and GRU
  • Deep Generative Models (Autoencoders and GANs)
  • Attention Mechanisms
Algonquin College — Ottawa, Canada
Summer 2025

Natural Language Processing (NLP)

Applied Undergraduate
  • Text preprocessing: tokenization, normalization, stopwords
  • Word representations: one-hot, TF–IDF, word embeddings (word2vec, GloVe)
  • Sequence modeling: RNNs, LSTM, GRU (concepts & limitations)
  • Transformers: attention mechanism, encoder/decoder overview
  • Language models: n-grams to pretrained LM (BERT/GPT high-level)
  • Common NLP tasks: classification, NER, POS tagging, QA, summarization
  • Evaluation metrics: BLEU, ROUGE, accuracy, F1
  • Practical pipelines: data labeling, fine-tuning pretrained models
  • Ethics & bias in NLP
Istanbul Gelişim University — Istanbul, Turkey
Fall & Winter 2022–2023

Data Mining

Graduate
  • Data preprocessing & cleaning
  • Exploratory data analysis and visualization
  • Supervised learning: decision trees, random forests
  • Unsupervised learning: clustering (k-means, hierarchical)
  • Association rules (Apriori) and pattern discovery
  • Evaluation: cross-validation, precision/recall, ROC
  • Scalability & basic big-data considerations

Object Oriented Programming

Undergraduate
  • OOP principles: encapsulation, inheritance, polymorphism
  • Classes, objects, methods, constructors/destructors
  • Design patterns (intro): factory, singleton, strategy
  • Exception handling and resource management
  • Unit testing basics and code organization
  • Practical programming assignments (language-specific)

AI for Business

Undergraduate / Applied
  • AI concepts and business use-cases
  • Data-driven decision making and KPIs
  • Supervised models for forecasting & classification
  • Intro to recommendation systems and customer analytics
  • Model deployment basics and production considerations
  • Ethics, privacy, and ROI of AI projects
Cyprus Science University — Kyrenia, Cyprus
Fall 2021

Neural Networks

Undergraduate / Introductory
  • Perceptron and multilayer networks
  • Activation functions and network architectures
  • Backpropagation derivation and practice
  • Training techniques and overfitting mitigation
  • Applications: classification, regression, simple vision tasks

Machine Learning

Graduate
  • Supervised learning: linear/logistic regression, SVM
  • Ensemble methods: bagging, boosting
  • Unsupervised learning: PCA, clustering
  • Model evaluation, bias–variance tradeoff
  • Feature engineering and pipelines
Islamic Azad University — Iran
Fall & Winter 2020

Information Retrieval

Undergraduate
  • Search architectures and indexing (inverted index)
  • Retrieval models: Boolean, vector space, BM25
  • Evaluation: precision, recall, MAP
  • Query processing, ranking, and basic NLP for IR
  • Web search basics and crawling/indexing concepts

Metaheuristic Algorithms

Graduate
  • Genetic algorithms and evolutionary strategies
  • Simulated annealing, tabu search
  • Particle swarm optimization
  • Applications to combinatorial & continuous optimization
  • Parameter tuning and hybrid strategies

Data Mining

Graduate
  • (See Data Mining above) — core topics tailored for local curriculum
  • Practical exercises: data preprocessing and pattern extraction