Machine Learning in Nepal
Machine Learning in Nepal is fastly growing in the field of education, government, and corporate business. Different companies in Nepal develop an AI system, used a machine learning method to complete their daily task. Nepal is currently in the second phase of AI(Artificial Intelligence). The second phase means the data collection phase. Private and Government sector is collecting data from users. This data are used to make a business decision governmental benefit and AI development.
What is Machine Learning?
Machine Learning is the process of making intelligent machines without an explicit program. The Machine Learning process is the same as a newborn baby. Newly born baby is able to recognize thing by continuously learn through the pattern from data. As the same Machine Learning algorithm is also recognized a new pattern of data by analyzing previous data. Machine Learning Algorithm totally depends on data. Some AI systems are rule-based and some are automatically operated by recognizing patterns.
If you need to make a machine to recognize the number(handwritten digit) we need a large (millions) collection of the handwritten digit to feed into the machine learning algorithm. If you need to analyze the sentiment of sentences we need a large collection of natural data. These data to give the ability to make a pattern and recognized future data to the machine.
Why Machine Learning in Nepal?
Nepal is a developing country. Nepal is somehow collecting data from user in different sectors. Every sector in Nepal struggles to get some peak points. Manufacturers, Agriculture, Business, and Government sector need some advanced process to serve people. Machine Learning Technology helps to regulate the daily operations and make some recommendations to personnel. It helps to decrease corruption and increase transparency. Machine Learning helps to make better decisions. It is also useful for crisis management. A different perspective of analysis Nepal need sophisticated AI to operate each and every business.
The benefit of machine learning
In agriculture, the machine learning system helps to recommend seed by analyzing the current environment. The weather prediction system informs farmers about the weather condition. Different Deep Learning technology helps to recognize disease on plants by simply using a mobile camera. Machine Learning and Robotics Technology help in advanced agriculture.
Like American and European country AI helps to understand current environment by continuously analyzing past data. Robots are automatically analyzed plant health conditions and treat them accordingly.
In Nepal, different manufacturing companies operate daily in a different area of the country. Real Estate business, hydropower, road, and the building are operating a daily basis. Machine Learning algorithm is used to predict Real Estate price. Machine Learning is used to model 3D construction and many more. Simulating hydropower projects using data and make decisions about the developing environment really helps the engineer. AI also able to the analysis environment and social aspect of these projects.
Machine Learning in Business is used to make business decisions. It also helps to analyze customer behavior. Let’s take an example of Amazon. Amazon used machine learning technology to recommend the best product to the customer by analyzing customer behavior. 10% of the overall revenue of amazon is collect from recommendation. Alibaba in the other hand sell there 15% of overall product using intelligent system. Machine Learning in Nepal also helps in the Business Sector to make a better decision.
In Education, the Nepal government needs to take serious action. In the Education sector duplicate certification, unauthorized documenting is a serious problem in Nepal. Machine Learning algorithm used to recognized educational fraud and validity of documentation.
In Nepal, Machine Learning is also used in Music industries, the Medical sector, smart cities, Travel & Tourism, and culture development.
Government Sector and Machine Learning
The different governmental sectors to work for developing countries. Agriculture, Manufacturer, Education, and the Medical sector need to improve it’s working mechanism. Machine Learning regulate governmental financial transaction, document regularity, validity checking, transparency. Currently, the government sector collects user data in every sector (electricity, education, land, etc ) digitally through digital devices. So due to a different perspective, it’s important to machine learning in Nepal.
ML Suggestion for the Government of Nepal
- Collect user data
- Collaborate with IT company
- Make Brief AI strategy
- Invest in AI Development
- Collaborate with foreign AI research and development program
Machine Learning Wireframe
Python Programming Language / R programming Language
- Introduction to Machine Learning
- Supervised vs. Unsupervised Learning
- Installing Anaconda and Managing Environment
- Familiarization with Datasets
- Machine Learning Libraries(Numpy, Pandas, Matplotlib)
- Importing the Libraries
- Importing the Dataset
- Missing Data
- Splitting the Dataset into the Training set and Test set
- Feature Scaling
- Optimization and gradient descent
- Linear regression implementation
- Correlation Analysis and Feature Selection
- Evaluate Model Performance
- Multiple Regression and Feature Importance
- Variance Bias Trade-Off – Validation Curve
- Variance Bias Trade-Off – Learning Curve
Support Vector Machine
- Linear SVM Classification
- Polynomial Kernel
- Support Vector Regression
- Principal Component Analysis
- K-mean clustering
- Hierarchical Clustering
- Tic Toc Toe case study
- Sentiment Analysis of Nepali sentences
- Nepali Handwritten Digit Recognition
In Machine Learning supervised machine learning is the process of feed label data to the machine. During machine learning development we need data. If we use label data to analyze pattern this is known as supervised learning.
In the context of Nepal in Agriculture we can, we supervised learning method. Ministry of agriculture has huge data about farmers so we can use supervised machine learning methods. In weather forecasting we can also use the same learning method to predict future weather conditions.We can find data set about Nepal’s agriculture in opendata Nepal.
Example of supervised learning
This is really good. Positive
This example shows label data of sentiment analysis.
- SVM(Support Vector Machine)
- Naive Bayes
- Decision Tree
In Machine Learning Unsupervised Machine Learning is the process of feed unlabeled data to make model. During the model development algorithm is automatically find patterns on data. Unsupervised machine learning is helpful when we don’t know about the pattern of data. In Nepal, unsupervised machine learning is useful in different domains. Most of the data in Nepal has no pattern so we can use unsupervised machine learning.
- k-Mean Clustering
- Hierarchical Clustering
Machine Learning in the milestone of current scenarios. Machine Learning in Nepal helps governmental, private, and corporate businesses. It is able to done different decisions based on previous data and experiences like humans. We can explore the knowledge of AI all over the country.