Recommender Systems in Practice Crash Course

Welcome to the Recommender Systems in Practice Crash Course in collaboration with Applied Singularity!

As the amount of content on Netflix or products on Amazon increases, consumers are given significantly more choice. However, choosing a single item takes much longer and the process becomes less enjoyable, or even frustrating, for users. A powerful Recommendation System can help users by suggesting and highlighting items that they are most likely going to want or need. These systems also help the platform by ensuring that users complete the transaction or consume the content faster. As such, they are indispensable to organisations and are at the core of AI efforts for multiple companies today. 

Recommendation Systems are becoming increasingly sophisticated, incorporating an ensemble of algorithms from Machine Learning, NLP, time-series modelling, and more. In this 6-hour workshop, you will be exposed to the inner workings of Recommender algorithms, see real-world use cases from social media companies, e-commerce companies, streaming content providers, etc, and get hands-on experience building your own Recommender System!

Feel free to ask your questions via Machine learning hub on Global AI Hub Community.

You can check most frequently asked questions about this course.

Syllabus:

Day 1

Introduction to Recommender Systems

Types of Recommender Systems – Content-based, Collaborative-based & Hybrid systems

Important concepts related to Data Analysis & Preprocessing – NLP based concepts like Stemming, Lemmatization

Statistical concepts – Pearson Correlation, Cosine Similarity etc.

Day 2

Various approaches for building Recommender Systems – Matrix Factorization, SVD, Deep Learning etc.

Recommender Systems in Practice – Amazon, Netflix, LinkedIn, YouTube

Impact of Personalization & Recommendations

Can try this – Kaggle Challenge: A task is given based on a dataset. Time limit of 7 days.

Pros & Cons of Recommendation Systems

Applied Singularity Quiz

Day 3

Exploratory Data Analysis on various datasets

Hands-on – Content Based, Collaborative (and it’s varieties) Based & Hybrid Implementations

Brief about the latest advancements in this field, Quiz

Ethics in building Recommender Systems

Prerequisites for the Workshop:

Conceptual understanding of Deep Learning

Basics of NLP

Workshop Requirements:

Stable Internet connection

Access to Google Collab

Machine Learning Model Training and Deployment Crash Course

Welcome to our Machine Learning Model Training and Deployment Crash Course, we are glad to see you with us!

In this training, theoretical information such as how to build a data science project, which steps does it consist of, what are the data preparation processes are mentioned. Then, an end-to-end example machine learning project with a real-world data set is built using Python programming language and its outputs are analyzed and evaluated. The trained model, after all, will be open to the visitors to be used while being visualized on the web using the Python library called Streamlit. Streamlit is very easy to grasp yet efficient to display all the required output of the model based on the according argument inputs of the user. While not using a single line of web development, and using completely Python throughout the course, you will learn how to interact with the web and deploy your own custom models in the future!

Feel free to ask your questions via Machine learning hub on Global AI Hub Community.

You can check most frequently asked questions about this course.

Content:

Day 1

Theory

Steps of Machine Learning Project

Random Variables

• Data Preprocessing

Exploratory Data Analysis

Train-Test Split

Evaluation of Models

Hands-on

Building Machine Learning Project with Real Life Dataset

Day 2

Display text, data, charts, media, interactive widgets, code, progress bar and status

Control flow

Add widgets to sidebar

Lay out your app

Placeholders, help, and options

Final Project: “Custom trained model deployment and usage by the visitors/members”

Web-App Project

Welcome to our Web App Project workshop in collaboration with TeensInAI, we are glad to see you with us! 

In our Web App Project workshop, we will be covering essential topics like general ML Models, Model Deployment, FastAPI, and Postman. Also, we prepared unique contents using different worldwide resources to support your learning process!

During the workshop, we will create a simple web app and visualization of the ml model work outputs online.  After the workshop, you will improve your theoretical knowledge and coding skills.

Enroll our Web App Project workshop and start to learn how frameworks and libraries work in model deployment, API architecture base, and gain hands-on experience of the general process and the algorithms standing in the behind of the whole process in Python!

You can ask all your questions via Machine Learning Hub on Global AI Hub Community.

You can check most frequently asked questions about this course.

 

Content 

  • Introduction: Machine Learning Models & Usage
  • Frameworks & Libraries for Model Deployment
  • General Process and Algorithms
  • Introduction: FastAPI
  • API Architecture
  • Test with Postman

Data Science

Welcome to our Introduction to Data Science workshop in collaboration with TeensInAI, we are glad to see you with us! 

In our Introduction to Data Science workshop, we will be covering essential topics like Data Preprocessing, Visualization, Model Training, and Report of the Output. Also, we prepared unique contents using different worldwide resources to support your learning process!

During the workshop, we will evaluate a simple data set, then we will train a Machine Learning model with this data. After the workshop, you will improve your theoretical knowledge and coding skills.

Enroll in our Data Science workshop and start to learn how to prepare the data by cleaning and filling the nulls, detecting the outlier, feature scaling, visualizing the prepared data, and gain hands-on experience in choosing the right model and training in Python!

You can ask all your questions via Machine Learning Hub on Global AI Hub Community.

You can check most frequently asked questions about this course.

 

Content 

  • Gathering & Preparing Data
  • Pre-processing: Handling duplicated, missing, or null values
  • Outlier Detection, Feature Scaling
  • Feature Extraction, Feature Encoding
  • Training, Validation, Test
  • Data Analysis & Training a Model