Artificial intelligence and machine learning started much earlier, but in recent times, with the help of power computing and the availability of data, the implementation of models has shown better results. One important part of the performance of a model in machine learning is the data. There are many ways of retrieving and cleaning the data, and we can also see the relationship between the features in the data by plotting, exploratory data analysis. Also, we need to process it before it is used to train a model, feature engineering. After processing the data and before training a model with it, one can raise hypotheses about the relationship between features of the data and test them; this is called statistical hypothesis testing.
- A brief History of Modern AI and its applications: quick introduction to AI and Machine Learning and a brief history of the modern AI. We will also explore some of the current applications of AI and Machine Learning.
- Retrieving and cleaning Data: retrieve data from different sources, how to clean it to ensure its quality
- Exploratory data analysis and Feature Engineering: exploratory analysis to visually confirm it is ready for machine learning modeling by feature engineering and transformations.
- Inferential statistics and Hypothesis testing: useful definitions and simple examples that will help you get started creating hypothesis around your business problem and how to test them
Access to safe drinking-water is essential to health, a basic human right and a component of effective policy for health protection. This is important as a health and development issue at a national, regional and local level. In some regions, it has been shown that investments in water supply and sanitation can yield a net economic benefit, since the reductions in adverse health effects and health care costs outweigh the costs of undertaking the interventions. The water_potability.csv file contains water quality metrics for 3276 different water bodies.
Hypothesis:
- Null hypothesis (H0): There is no difference in the mean of a numerical variable (for example, pH, hardness, TDS, etc.) between drinking water and non-potable water.
- Alternative hypothesis (H1): There is a difference in the mean of the numerical variable between drinking water and non-drinking water.
Conclusion: Based on the p-value results, the key factors affecting water potability are:
- H: Critical for maintaining water safety.
- Chloramines: Essential for proper disinfection but harmful at high levels.
- Organic Carbon: Indicates the presence of natural organic matter, potentially harmful.
- Turbidity: High levels indicate suspended particles and microorganisms.
Other factors like Hardness, Solids, Sulfate, Conductivity, and Trihalomethanes are not statistically significant in determining potability. Focus on controlling pH, chloramines, organic carbon, and turbidity to ensure safe drinking water.