This patent discloses systems, methods and devices for prioritizing calendar events with artificial intelligence to resolve scheduling conflicts. When a request to schedule a new event clashes with an existing one, the system compares their “event priority scores,” generated by a statistical machine learning model considering various factors. If the new event’s score is higher, a selectable option to replace the conflicting calendar event with the new calendar event may be presented.
Main Themes and Important Ideas/Facts:
- Problem Addressed: The patent addresses the common issue of electronic calendars becoming cluttered with low-priority events, requiring manual intervention to reschedule when higher-priority events need to be accommodated.
- Core Solution: AI-Powered Prioritization: The central idea is to use artificial intelligence, specifically statistical machine learning, to automatically prioritize calendar events and suggest or perform rescheduling based on these priorities.
- Event Priority Score Generation: A key component is the generation of an “event priority score” for each event. This score is determined by applying a statistical machine learning model to a “plurality of factors” associated with both the new and the conflicting events.
- Factors Influencing Priority Scores: These factors can be broadly categorized into:
- Event Parameters: These relate to the event itself, such as duration, time and day, location, type of attendance (in-person/electronic), agenda, booking time, and event history (e.g., number of reschedules).
- Attendee Attributes: These relate to the individuals involved in the event, such as seniority, organizational title, office location, tenure, salary range, and performance review data. User opt-in is suggested for accessing this information.
- Conflict Resolution Mechanism: When a new event conflicts, its priority score is compared to the existing event(s). If the new event has a higher score, the user may be presented with an option to replace the lower-priority event.
- Machine Learning Model Details: The patent specifies the use of a “statistical machine learning model,” potentially a “feature selection model” like a “Bayesian model.” In some cases, a “t-test” might be applied before the Bayesian model.
- Confidence Score: A “confidence score” may be calculated to represent the likelihood that a user will prefer the new event over the conflicting one. This score might need to meet a certain threshold for the replacement option to be presented or for automatic rescheduling to occur.
- User Feedback and Model Training: The system incorporates a feedback loop. When a user accepts or rejects the suggested replacement, this feedback is used to train and refine the machine learning model, improving its accuracy over time. Unsupervised clustering may be used to improve future suggestions based on similar events.
- Automation and User Control: The system can be configured to either automatically reschedule lower-priority conflicting events or to present users with selectable options before any changes are made.
- Technical Advantages: The patent highlights reduced processing costs (fewer messages exchanged for scheduling) and minimized memory costs (conflicting meetings are moved rather than remaining). The feedback mechanism aims to improve the efficiency of the prioritization over time.
- Integration with Digital Assistants: The system can interact with users through natural language requests processed by a digital assistant.