RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation (CIKM'17)

Last update: Feb 10, 2022

Overview

RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation

This is the implementation of RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation.

Code

To run our code, please use the following commands:

g++ RATE.cpp -o RATE -std=c++11
./RATE [Training File] [Test File] [L, optional, default = 30] [T, optional, default = 1]

For example,

g++ RATE.cpp -o RATE -std=c++11
./RATE Dataset/train.txt Dataset/test.txt 40 1

The prediction results will be in ./result.txt (the first row is the classification result). Then you can run

python eval.py

to obtain evaluation metrics.

Dataset

We release the Europe dataset (Dataset/data.json), where each line is a json file with tweet text and metadata. Due to privacy issues, we have anonymized the whole dataset by representing each word/feature as an integer. An example is shown below.

{ 
   "label":0,
   "language":"3",
   "timezone":"5",
   "offset":"7",
   "userlang":"5",
   "latitude":"36.8901",
   "longitude":"30.6809",
   "text":"3332 2608 29"
}

Given the json file, one can run

cd Dataset/
python preprocess.py

to get training and testing data (Dataset/train.txt and Dataset/test.txt).

Result

Method	Micro-F1 (Acc)	Macro-F1	Mean Distance Error (km)	[email protected]
RATE	0.8905	0.5230	365.16	0.4315

Citation

@inproceedings{zhang2017rate,
  title={RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation},
  author={Zhang, Yu and Wei, Wei and Huang, Binxuan and Carley, Kathleen M and Zhang, Yan},
  booktitle={Proceedings of the 2017 ACM on Conference on Information and Knowledge Management},
  pages={2423--2426},
  year={2017},
  organization={ACM}
}

RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation (CIKM'17)

Related tags

Overview

RATE: Overcoming Noise and Sparsity of Textual Features in Real-Time Location Estimation

Code

Dataset

Result

Citation

Owner

Yu Zhang

Source code of article "Towards Toxic and Narcotic Medication Detection with Rotated Object Detector"

Focal Loss for Dense Rotation Object Detection

The GitHub repository for the paper: “Time Series is a Special Sequence: Forecasting with Sample Convolution and Interaction“.

MODALS: Modality-agnostic Automated Data Augmentation in the Latent Space

This repo is duplication of jwyang/faster-rcnn.pytorch

TensorFlow implementation of original paper : https://github.com/hszhao/PSPNet

基于YoloX目标检测+DeepSort算法实现多目标追踪Baseline

This repo provides a demo for the CVPR 2021 paper "A Fourier-based Framework for Domain Generalization" on the PACS dataset.

[CVPR22] Official codebase of Semantic Segmentation by Early Region Proxy.

The official PyTorch implementation for NCSNv2 (NeurIPS 2020)

This is the code of paper ``Contrastive Coding for Active Learning under Class Distribution Mismatch'' with python.

Computational modelling of ray propagation through optical elements using the principles of geometric optics (Ray Tracer)

A modular active learning framework for Python

Re-implementation of 'Grokking: Generalization beyond overfitting on small algorithmic datasets'

Code for Paper "Evidential Softmax for Sparse MultimodalDistributions in Deep Generative Models"

Code of the paper "Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition"

This package contains deep learning models and related scripts for RoseTTAFold

DeepCO3: Deep Instance Co-segmentation by Co-peak Search and Co-saliency

Adversarial Self-Defense for Cycle-Consistent GANs

Improving Contrastive Learning by Visualizing Feature Transformation, ICCV 2021 Oral