Hello, I'm
Walid Iguider.

About.

MY MISSION IS TO Lead, DESIGN & DEVELOP Innovative AI/ML products

Walid Iguider, Ph.D. is a senior data scientist at Stackhouse (Spindox S.P.A. company) . He has about a decade of total experience, with more than 4 years in Data Science, more than 5 years in Software Engineering and 3 years of Agile Project Management. As a Senior Data Scientist, Dr. Iguider helps business owners reach their goals using AI based technologies. He leads, designs & develops innovative AI/ML products, based on Recommender Systems, Sentiment Analysis, Decision support systems...

Prior to joining Stackhouse, Dr. Iguider was a senior data scientist at Athlos. Earlier, he was a doctoral researcher at the Artificial Intelligence and Big Data Laboratory and collaborating with Abinsula Srl and the Data Science and Big Data Analytics research group of Eurecat.

Dr. Iguider has a master's degree in Software Quality from Sidi Mohamed Ben Abdellah University of Fez and completed his doctoral program in Computer Science, with a focus on Data Science, at the University of Cagliari. In particular, the goal of his Ph.D. was to levrage Machine Learning to support decision-making in sports remote coaching plateforms.

His experience includes collaborating with national and international companies in Emirates, France, Italy, Morocco, and Spain.
He is fluent in Arabic, English, French, and Italian.

Open source projects.

Active Learning for Text Classification in Python and Small-Text

This project consists of a notebook illustrating a simple example of how to perform Active Learning for text classification in python, leveraging on small-text (a library that provides state-of-the-art Active Learning for Text Classification) using sklearn models and the rotten tomatoes dataset. The main active learning loop queries the unlabeled pool and thereby decides which documents are labeled next. We then provide the labels for those documents and the active learner retrains the model. After each query, we evaluate the current model against the test set and save the result.

learn more...

Open source projects.

Convolutional Autoencoders for Image Reconstruction in Python and Keras

This project introduces an example of a convolutional (variational) autoencoder that reads an input image, encodes the image into a lower dimensional latent representation, then decodes the latent representation to reconstruct the imput image. The autoencoder is implemented using the Keras, and it is based on convolutional neural networks leveraging Conv2D layers in the encoder, and Conv2DTranspose layers in the decoder. The autoencoder is trained using the Fashion-MNIST dataset. Each image in this dataset is 28x28 pixels. For this reason, the input shape of the encoder was set to (28, 28, 1) as well as for the output shape of the decoder.

learn more...

Open source projects.

An Ensemble Deep Learning Approach to Classify Scoliosis and Healthy Subjects In Python and Keras

In this project we constructed and trained an Ensemble Neural Network to classify scoliosis and healthy subjects. The model predicts the probability that a subject suffers from AIS basd on the VRS data. The Ensemble Neural Network is implemented using the Keras, and it is averaging the prediction of 16 multi-layer perceptron (MLP) neural networks. Each neural network is made of two hidden (Dense) layers and each layer is composed of 64 neurones. The Ensemble Neural Network performs quite well achieving a balanced accuracy over 86%.

learn more...