INTELLIGENT SYSTEM DEVELOPMENT
EXERCISE 5
Tuần sau (23/09---26/09) đánh giá điểm
=================================================
Exercise 5.1 Predicting, classifying classes based on data with weight, height and jobs:
Step 1 (2d)
o Generate dataset with 2000 persons including height (m), weight (kg), job (30 jobs).
Example jobs: doctor, dentist, engineer…
o Based on calculate the Body Mass Index (BMI) of this group
of people. The formula to calculate BMI is as follows:
Divide the weight in kilograms (kg) by the height in meters (m)
- Divide the answer by the height again
Using the BMI, you can classify a person as healthy, overweight, or underweight using
the following categories:
- Underweight if BMI < 18.5
- Overweight if BMI > 25
- Normal weight if 18.5 <= BMI <= 25
o Using models linear regression and logistic regression to predict weight for certain
height and job types.
Step 2 (2d)
o Compare and evaluate
▪ Evaluate and show plots of accuracy with metrics MAE, RMSE when comparing
Linear Regression and Logistic Regression to predict weight. Show plots in one
diagram
▪ Show distribution of persons with heights, weights and jobs. Discussion
o Implement interfaces for users entering the height and job, and show weight:
▪ GUI
▪ Web interface
▪ Mobile (Android, iOS)
Step 3 (2d):
o Build deep learning models CNN, RNN with tensorflow (NO KERAS) for predicting weight
similar to step 2. Show in the interface to user selecting technique
Step 4:
o Implement a chatbot interface to give health-related recommendations (e.g., weight
gain or loss advice) based on predictions.
o Chatbot also gives advices to use kinds of foods w.r.t. fat or thin diseases…
Exercise 5.2
Applying Exercise 1 for predicting examination marks in the previous Exercise for a set of data for 2000
students
Exercise 5.3
Refer to Exercise 3.7 for 2000 students. You demand ChatGPT generate 2000 students with attributes
given in Exercise 3.7
