🛡️ Privacy Notice and Data Usage Statement

This page explains how "Waters of Xinyi" processes your data, and the rights you have.

💧 Project Overview

"信水義河" is an interactive hydrological guide map project developed as part of the "Smart City and Digital Democracy" course at the NTU Graduate Institute of Building and Planning, in collaboration with Taipei Xinyi Community College. The project uses digital tools to help the public explore the hydrological history and environmental change of Taipei's Xinyi District, and to foster public participation through a feedback mechanism.

📋 Scope of Data Collection

The following are the categories of user data that this project may collect:

Data Type	Collection Method	Sensitivity
Feedback text	Actively entered by the user	🟢 Low
Photos	Actively uploaded by the user (may include faces)	🔴 High
GPS coordinates	Automatically extracted from EXIF data + user map markers	🟡 Medium
Camera device model and date/time	Automatically extracted from EXIF data	🟢 Low

🚫 Data Explicitly Not Collected

To protect your privacy, the following data is never collected or stored by this project:

👤User name, email address, or phone number

🍪Browser cookies or tracking identifiers

🌐IP address

🎙️Speech recording files

📊Browsing history or behavioural analytics

🤖 AI System Usage

⚠️ Important: All AI features are opt-in (user-initiated), not automatic. Users must manually activate each feature, and may preview, edit, or reject any AI-generated output at any time.

Text Processing

Gemini 2.5 Flash — AI-Assisted Text Editing

Once manually triggered by the user, AI assists in editing the feedback text. Users may preview, edit, or reject the AI-modified content.

Image Analysis

Gemini 2.5 Flash — Image Description

After uploading a photo, users may actively trigger AI to generate descriptive text. The generated result can be edited by the user.

Speech-to-Text

Web Speech API — Speech Recognition

Speech recognition is processed entirely locally in the browser; no recording files are stored. The recognition result is displayed in real time for the user to confirm.

🔗 Third-Party Data Transfers

This project uses the following third-party services; the scope of data transferred and its purpose are shown in the table below:

Third-Party Service	Data Transferred	Purpose
Google Gemini API	Text / compressed photos	AI inference (text editing, image description)
Google Apps Script	Form data / photos	Data storage
Google Drive	Compressed photos	File storage
Google Earth Engine	No user data	Satellite imagery processing
Web Speech API	Speech audio stream	Speech recognition (local, in-browser)

🗄️ Data Retention and Deletion

💾

Feedback text / photos / EXIF data

Retained permanently, until manually deleted by an administrator

🎙️

Speech recordings

Not stored — speech recognition is processed locally in the browser only

📝

Original text before AI editing

Not stored — only the final version confirmed by the user is retained

👁️ Content Moderation

✅

Manual Review System

All feedback submitted by users defaults to "Pending Review" (pending) status, and requires manual approval by an administrator before it can be publicly displayed on the map.

🤖

No Automated AI Review

This project does not use any AI to automatically decide whether user-submitted content is published. All publication decisions are made by humans.

✊ User Rights

When using this project, you have the following rights:

🤖Freely choose whether to use AI features

📷Freely choose whether to upload photos

📍Freely choose whether to share EXIF GPS data

👀Preview all submitted content before uploading

📬 Contact

If you have any questions about this privacy notice, or need to exercise the user rights described above, please contact:

NTU Graduate Institute of Building and Planning

"Smart City and Digital Democracy" Course Team

Last updated: 2026-06-03

← Back to map