This repository contains the code used in our paper: TASKS: A Real-time Query System for Instant Error-Tolerant Spatial Keyword Queries on Road Networks. TASKS is a real-time query system for instant error-tolerant spatial keyword queries on road networks, which consists of an index component, a query component, an update component, and a route component.
a 64-bit Linux-based OS; a windows OS
We use a publicly available real-world road networks of New York as an example, which is obtained from ...
The data is shown in folder demo-backend/TASK/datasets/NY
.
-
Step 1: Push the demo-backend part code to the linux server
-
Step 2: Move to directory
demo/demo-backend
and run flask server, through which the data interaction between the front-end interface and the back-end TASK is realizedpython demo_flask.py
-
Step 3: Move to directory
demo/demo-backend/TASK
and run back-end TASK derived from TASKmake go
-
Step 4: Open the front-end interface
/demo-frontend/TASKS-demo.html
at Windows to use TASKS
- cmake
- g++
- OpenMP
- Boost
- flask