Fresh Air Match: local AI that gets you outside without tracking you

Описание и применение

Приложение Fresh Air Match использует локальную модель MiniLM для быстрого и приватного предложения трёх коротких outdoor‑активностей, учитывая ограничения пользователя.

Задача

Users want quick, privacy‑preserving outdoor activity suggestions based on mood and constraints.

Результат

The app delivers three curated outdoor activities within user constraints, preserving privacy and ensuring deterministic safety filtering.

Рабочий процесс

• Collect user constraints (time, weather, company, energy, access, free‑text mood).
• Apply deterministic safety/access filters to a curated catalogue of 16 activities.
• Run local MiniLM embeddings on the free‑text note.
• Rank remaining activities by cosine similarity.
• Return top three activities with time bounds, plan, bring list, and safety reminder.

Доказательства

Fresh Air Match turns a short description of your mood into three small outdoor activities you can actually do.

Источник

It filters a curated catalogue of 16 activities using the hard constraints first. Only the eligible activities reach the semantic-ranking stage.

Источник

A local sentence-transformers/all-MiniLM-L6-v2 ONNX model then compares the note with those activities and returns the three closest matches.

Источник

The model runs locally through ONNX Runtime. Preference text is not sent to a hosted model, and the app asks for no precise location.

Источник

Semantic similarity never overrides a hard constraint. The server first removes activities that do not fit the requested time, company, weather, setting, energy, or access requirements.

Источник

I ran ten unit and integration tests covering: input validation and length limits; catalogue integrity; time, weather, company, setting, energy, and accessibility filters; semantic ranking; privacy and safety statements; and real ONNX inference when the model artifacts are present.

Источник