handiism / kalorize-ml

Machine Learning

Food Recommendation From Calories Needed

Calories are nutrients that affect the human body in terms of energy and weight management. They are a unit of measurement for the amount of energy provided by food and beverages. Calories are essential for the body to perform its basic functions, such as breathing, circulating blood, and repairing cells. The number of calories needed varies from person to person, depending on factors like age, gender, weight, height, and activity level. Consuming more calories than the body needs can lead to weight gain, while consuming fewer calories can result in weight loss. Different types of food contain different amounts of calories. For example, foods that are high in fat and sugar tend to be higher in calories. Regular physical activity can help burn calories and maintain a healthy weight. Monitoring calorie intake can be helpful for individuals looking to manage their weight or make dietary changes.

We created food recommendation from nutrient that needed. From questionnare that we given and machine learning, give food recommendation for three times, breakfast, lunch, and dinner. We use Nearest Neighbors algorithm. The algorithm find similarity from the nutrient of food items in the database and the nutrient needs specified in the questionnaire. The Nearest Neighbors algorithm analyzes the nutrient profiles of various foods and identifies the ones that closely match the individual's requirements. It takes into account the quantities of different nutrients, such as protein, carbohydrates, fats, vitamins, and minerals. The algorithm ranks the food options based on their similarity to the desired nutrient profile.

Dataset

• Kaggle

References

• Nutrition Diet Recommendation System

Height, Weight And Gender Prediction From Face

Predicting height, weight, and gender from a face alone is a complex task that relies on various factors and approaches. While facial features can provide some clues about these attributes, they are not solely determinative, and other factors such as genetics, body composition, and lifestyle also play significant roles.

Height prediction from a face is challenging as height is influenced by genetic factors, bone structure, and overall body proportions. While certain facial characteristics may correlate with height to some extent, accurately predicting height solely from facial features is highly challenging and not commonly done.

Weight prediction from a face is also a difficult task. Facial features can provide some insights into body fat distribution, such as fullness of cheeks or the presence of facial fat deposits, but they cannot accurately determine an individual's weight. Weight is influenced by multiple factors, including body composition, muscle mass, and overall lifestyle.

Gender prediction from a face has received more attention in computer vision research. Facial characteristics, such as jawline shape, eyebrow thickness, and lip curvature, can provide some indications of gender. Machine learning models trained on large datasets of faces labeled with gender can learn patterns and make gender predictions based on these facial features. However, it's important to note that gender prediction from a face alone is not always accurate and can be influenced by factors such as grooming, hairstyle, and cultural variations.

Dataset

• Face's Dataset: 1544

References

• Keras-VGGFace

The Worker

Member	Student ID	Path
Iman Abdurrahman	M306DSX0337	Machine Learning
Ilham Aziz Nur Mahmudin	M169DSX1431	Machine Learning