Artificial Intelligence and Analytics Plan for Provisional Patent Application in Consumer Health

Project Description

We are a consumer health startup that helps people build personalized healthy lifestyles. We need a data expert to help us describe the artificial intelligence and analytics plan for a provisional patent application of our information technology system.

We are looking for a data expert with the following experiences.

1. Broad range of artificial intelligence including:

• Prioritization algorithms
• Recommendation algorithms
• Adaptive learning algorithms
• Machine learning algorithms

2. User Profile Analytics

• Performance metrics
• Scoring metrics

3. Describing AI models, algorithms, and analytics for provisional and utility patent applications

System Description

The function of the system is to enable users with a given health goal (e.g. weight loss, asthma, etc.) to discover consumer health solutions (products, services and behaviors) that might be effective for them. Instead of adhering to a generic program say for weight loss, users will be provided with tools to intelligently personalize a repertoire of solutions.

As part of this process, data collected about the user by various means will be used by an AI system to prioritize and recommend assessments, challenges, and solutions.

The process supported by the system, and the algorithms and analytics that enable this process are the core of the provisional patent application.

Similar Systems

Netflix, Pandora, and Khan Academy are example systems that illustrate the type of algorithms we are looking to apply. Though they share some similarities to the present system in other domains, they are substantially different in process and function.

In each of these systems, user data is gathered to create a profile, which in turn is the basis for prioritized recommendations from a database of movies, music, or educational modules. Based on user choices and outcomes, adaptive learning algorithms are applied by at least one of these systems. In iterative fashion, the best experiment, or recommendation, is identified for a given user to yield the most information for determining the best experiment in the next iteration.

Deliverable

AI and analytics plan for a provisional patent application

Requirements

1. Provisional patent applications require:

• Enough detail to clearly describe the claimed invention
• Claim a broader range of AI approaches than will ultimately be implemented or applied for in the follow utility patent application. Claiming this broader range in the provisional patent application is important, as it sets the date for the scope of our claim in the follow up utility patent application.

2. Prioritization engine and Recommendation engine description The application requires describing Data in/Transformation of data/Data out. We already have a list of data elements that can be used as input.

• We need help with determining the range of AI models, and how they would be applied to the data elements to prioritize and recommend assessments, challenges, and solutions. We think adaptive learning algorithms are a strong candidate, but as mentioned, we are looking for a range of AI models.
• The algorithms do not need to be made in real time, but rather on a nightly/daily basis.

3. User Profile Analytics As users repeat the process of discovering solutions, they will collect those solutions in their profile. User metrics need to be calculated to improve the next round of prioritizations and recommendations. We have a rough draft of Performance metrics and Scoring focused on determining effectiveness and sustainability of the set of solutions collected and how well they fit other user profile data.

• We need help refining and describing these Performances Metrics and Scoring and how they would be used by the prioritization and recommendation engines for the application.

4. Integration of analytics plan into existing draft of provisional patient application Our patent writer has already begun a draft of the application, and our system architect has created the system architecture. We have all the pieces required for the application except for the AI/analytics plan.

• The AI/analytics plan needs to be integrated into the application. You will be able to work with both the patent writer and the system architect.

5. Sign a Non-Disclosure Agreement

Work done so far:

• Detailed system description of process, user flow and features
• System architecture
• Data elements defined, though no data has yet been collected
• Basic examples of graphical user interface

Reference patents identified The AI/analytics plan is the remaining piece of the provisional patent application.