Omdena Chapter Page: Srilanka

Omdena Sri Lanka Chapter - Omdena Chapters

Welcome to the Sri Lanka Chapter!

Upcoming Projects

 

Project Starts: 24.06.2022

Duration: 4 Weeks

Sri Lanka Chapter – Implementation of AI powered chatbot to help with the general Mental-hygiene of the Sri Lankan community (starts 24.06.2022)

Background

 

 Mental-hygiene is one of the prominent  parameters in human life to maintain a balanced lifestyle.Mental health is related to both internal and external factors. Social implications are a very crucial factor which leads to drain human psychology. Amount that a person could spend for  basic needs reflects the status of the economy of the country. Most social research has proven that there is a correlation between economic status and mental wellness of a person. Above status would reflect with words that they utter or write.  As humans, we must maintain our mental health as well as physical health for a long life.


Problem

Sri Lanka is a third world country that is located in south Asia near to the equator. This is one of the most beautiful countries in the world enriched with various resources. Though Sri Lanka has multiple natural  resources, some unlucky factors leads this country towards an  economic crisis. Due to the fact that  Sri Lankan people are facing critical issues  in satisfying basic needs. Hence,  it is very important to take care of each other until Sri Lankans re-state the economy . Purpose of the challenge is to develop an AI based mental health care Chatbot which helps a person maintain his or her mental health.


Project Goals 

  •  Sufficient understanding on Deep learning in NLP

  • Creating a Chabot based on Mental health

  • Understanding of human behavior in different states

  • Developing a model or application which can detect and predict the human behavior in economic crisis

 

Learning Outcomes:

– Data Pre-processing and EDA

–  Industrial use cases of Natural Language Processing(NLP)

–  Mental-Hygine behavior analysis

–  End user behavior analysis in NLP

 

 

The Tasks & Timeline

 

 Week 1  Week 2  Week 3  Week 4

 

Data Collection and Data preparation

 

Data Cleaning and EDA

Model Creation and Application Creation Model Creation and Application Creation

 

Completed Project(s)

Using AI for Improving Quality of Green Tea Leaves in Sri Lanka

Background

Sri Lanka is one of the leading Tea producers in the world. The tea industry contributes to more than 5% of the national GDP. The tea manufacturing process is a pipeline process and the total output of the tea production relies entirely on the quality of green leaves received at the tea factory. Unfortunately, a significant amount (10% – 40%) of total leaves get damaged during the collection, packing, and transportation process.

The value chain starts with the tea leaves being plucked by humans to be stored in a basket or a special sack. Once tea leaves are plucked, the leaves are collected and packed into separate linen sacks and delivered to a factory by lorries. At the factory, the leaves get weighed by a supervisor and scattered to the special turf to start the weathering and fermentation processes. 

The initial quality of tea leaves is checked at the factory through which the tea leaves get classified into 3 main categories:

  1. Best
  2. Below Best
  3. Poor 

The quality checking process is conducted by a technical manager or a tea inspector appointed by the Sri Lanka Tea Board. Good quality tea must have a minimum of 60% of ‘Best’ tea leaf classifications.

The quality check is a lengthy and time-consuming process that needs to be completed prior to every tea manufacturing cycle. The price will be decided based on the above quality parameters.

The Problem

Identifying the type of tea leaves at the initial step of the production process using cutting-edge technologies to improve efficiency. 

Improve the quality of the tea using state of art technologies and prepare a transparent communication platform to share the information among factory owners, tea inspectors, and green leaf collectors. stakeholders in the tea value chain system. 

This tea manufacturing process can be exponentially expedited with the help of a Deep Learning based image classification solution. This solution will help to increase the efficiency, transparency, and profitability of the tea value chain process.  

The Project Goals:

1. Collect images of tea green leaves  from tea factories located around main tea regions 

2. Tag images based on the specified regions and the category of tea

3. Train the simple classification model to identify basic tea quality categories. 

4. Train complete model for each region

5. Develop mobile applications for the classification Tea qualities.

The Learning Outcomes

– Data Pre-processing and EDA

– Developing and Deploying Dashboards

– Image Classification algorithm Usage

– Computer vision and model Building

Using AI for Improving Quality of Green Tea Leaves in Sri Lanka (Phase-2)

Background

Sri Lanka is one of the top Tea manufacturers in the world and it holds the rank No 1 in the black tea exporter category throughout the past 2 decades. The tea industry contributes more than 5% to the national GDP. The tea manufacturing process is an interlinked process and the final output of the tea production depends entirely on the quality of green leaves received from Green Tea dealers  to the tea factory. 

Unfortunately, a significant amount (10% – 40%) of total leaves get damaged during the plucking, collection, packing and transportation process. To mitigate the issue of receiving poor leaves to the factory, we have decided to identify tea leaves using AI with an image processing method during the initiation phase of this challenge. Under the Machine Learning  life cycle following phases have been completed successfully and we will do further improvements during the 2nd phase of the project. 

FInalized Tasks under Machine life cycle

Expected UI design under the final Solution

Remaining steps will be started from the model which we have already built during the 1st phase and embed the model to a mobile application , so that users could operate the ML based classification model without any inconvenience.

The Problem

Though we develop the ML based model with higher accuracy, it’s pointless if we are unable to deploy the application on a portable and versatile device, so that real world users can classify tea leaves on their own, without any inconvenience. Therefore, it’s mandatory to identify existing barriers and difficulties prior to deploying a mobile application with the MLOps Cycle. Image capturing, uploading, preprocessing and training model has to be a recursive process. This might help to improve the accuracy of the model incrementally and give better results to end users.  So, deploying a fully fledged mobile application will be focused on this phase.  

The Project Goals:

1. Develop mobile applications for classification Tea qualities, based on main tea regions.

2. Using edge computing and cloud development for mobile applications

The Tasks & Timeline

 

 Week 5  Week 6  Week 7  Week 8

 

Product Design – Model Management & Serving

 

Model Management & Serving

 

Documentation & Reporting

 

Final Solution Alignment

Completed Projects

Srilanka Chapter Leads

Indika Hiran Wijesinghe

Indika Hiran Wijesinghe is an IT professional, an entrepreneur, Data Science evangelist who has been helping public sector institutes to adopt AI/ML-based analytics to shape their IT strategy. Also, he has been a mentor for numerous ICT startups where he advised in selecting correct DSML products that go with business strategy. Currently, he holds two masters including Big Data Analytics from Robert Gordon University. He is passionate about exploring and learning novel AI/ML technologies and applying them for the betterment of public sector Institutes and society at large

Vidura Bandara Wijekoon

Chief Operating Officer @Trinet Innovations|Data Analyst|Junior Machine learning engineer @Omdena|AI, Machine learning, Data Science enthusiastic|Bsc Electrical and information Engineering Undergraduate at University Of Ruhuna

 

 

 

 

 

 

 

The Learning Outcomes

1. Develop a portable device solution 

2. Integrate the mobile device solution with the trained models

3. Apply the solution for the real environment and fine tuning the application. 

4. Understanding of MLOps Life cycle

5. Understanding of server and client side of the application