Skip to content

Latest commit

 

History

History
172 lines (102 loc) · 5.23 KB

b_use_cases.md

File metadata and controls

172 lines (102 loc) · 5.23 KB

Project Apero Banner

Use Cases

Identifying and detailing use cases ensures that the Project Apero addresses real user needs effectively. Below are five distinct use cases that illustrate the application's functionality and value to its users.


Use Case Diagram

To visualize the interaction between actors and use cases in the system, we have created a use case diagram.

Use Case Diagram


Use Case 1: Personalized Health Recommendations

Description

A user inputs personal health metrics such as age, gender, BMI, blood glucose levels, and other relevant data. The system analyzes this information to provide tailored health recommendations aimed at reducing the risk of stroke.

Actors

  • Health-Conscious Helen
  • Senior Citizen Sam

Preconditions

  • User has access to the application.
  • User has input accurate and complete health data.

Flow of Events

  1. User navigates to the "Personalized Recommendations" page.
  2. User enters personal health metrics into the input widgets.
  3. System processes the data and generates personalized recommendations.
  4. Recommendations are displayed to the user with actionable steps.

Postconditions

  • User receives customized health advice.
  • User can choose to follow or seek further assistance based on recommendations.

Use Case 2: Data Analysis Exploration

Description

A user explores the data analysis section to view visualizations and insights about health data. The dashboard provides interactive charts and statistics, enabling the user to uncover trends and correlations.

Actors

  • Data Analyst David
  • Health-Conscious Helen

Preconditions

  • Data is loaded and preprocessed in the system.
  • User has access to the "Data Analysis" page.

Flow of Events

  1. User selects "Data Analysis" from the sidebar.
  2. User applies filters using interactive widgets to refine the data view.
  3. System displays updated visualizations based on applied filters.
  4. User explores various charts and statistical summaries to gain insights.

Postconditions

  • User gains a comprehensive understanding of the data.
  • User identifies key factors influencing health outcomes.

Use Case 3: Chatbot Assistance for Health Queries

Description

Users interact with the integrated Rasa chatbot to ask questions related to their health data, seek recommendations, or request data analysis summaries. The chatbot provides immediate and context-aware responses.

Actors

  • Health-Conscious Helen
  • Senior Citizen Sam
  • Data Analyst David

Preconditions

  • Chatbot is properly configured and connected.
  • User has a basic understanding of interacting with chatbots.

Flow of Events

  1. User navigates to the "Chatbot" page.
  2. User initiates a conversation by typing a query.
  3. Chatbot processes the input and retrieves relevant information.
  4. Chatbot responds with appropriate answers or recommendations.
  5. User can continue the conversation or end the session.

Postconditions

  • User receives timely and relevant information.
  • Enhanced user engagement through conversational interaction.

Use Case 4: Model Evaluation and Selection

Description

Data Analyst David evaluates multiple machine learning models to determine the most effective one for generating health recommendations. The system provides detailed metrics and comparison tools to aid in model selection.

Actors

  • Data Analyst David

Preconditions

  • Multiple models have been trained and saved.
  • Evaluation metrics are available.

Flow of Events

  1. User accesses the "Model Evaluation" section.
  2. User reviews evaluation metrics displayed for each model.
  3. User compares performance metrics of different models.
  4. User selects the best-performing model for deployment.
  5. System confirms the selected model and updates configurations accordingly.

Postconditions

  • Optimal model is chosen based on performance metrics.
  • System is configured to use the selected model for recommendations.

Use Case 5: Data Augmentation and Outlier Management

Description

The system automatically augments the dataset with synthetic data to enhance model training and identifies outliers to maintain data integrity. This process ensures robust and reliable recommendations.

Actors

  • Data Analyst David

Preconditions

  • Original dataset is loaded into the system.
  • Data augmentation and outlier detection algorithms are implemented.

Flow of Events

  1. User navigates to the "Data Augmentation" section.
  2. System processes the data to identify and remove outliers.
  3. System generates synthetic data to augment the original dataset.
  4. User reviews the updated dataset through visualizations.
  5. Enhanced dataset is used for training more accurate models.

Postconditions

  • Dataset is free from significant outliers.
  • Augmented data improves model performance and reliability.

Note: The use case diagram provides a visual representation of how the actors interact with the system's functionalities, emphasizing the roles and relationships within the application.