lixus7 / time-series-works-conferences Goto Github PK

Hi,

Here is the GitHub repo for FuSAGNet (KDD22):

Thanks!

Thanks for your great work on collecting time-series papers.

For the paper "FluxEV: A Fast and Effective Unsupervised Framework for Time-Series Anomaly Detection", you can find the code at https://github.com/jlidw/FluxEV. Could you please update it? Thanks very much!

Hi,

Thanks for a great work.
I just made a PR with some minor issues I noticed #9, but there was a larger change I thought of.

Currently, the CCF ranks of the conferences are written for each publication which seems unnecessary. Instead, I think you can just move this information to be under the "Conferences" title. In this way, the information regarding the ranks of the conferences (i.e., what the ranks are, conference average quality, etc.) could also be moved to "Conferences".

1. Multi-Granularity Residual Learning with Confidence Estimation for Time Series Prediction The code repository for this paper was deleted；
2. EXIT: Extrapolation and Interpolation-based Neural Controlled Differential Equations for Time-series Classification and Forecasting
   code https://github.com/sheoyon-jhin/EXIT
3. Paper name of CAMul is wrong. [CAMul EXIT: Extrapolation and Interpolation-based Neural Controlled Differential Equations for Time-series Classification and Forecasting]
4. RETE: Retrieval-Enhanced Temporal Event Forecasting on Unified Query Product Evolutionary Graph
   code https://github.com/DiMarzioBian/RETE_TheWebConf
5. TLogic: Temporal Logical Rules for Explainable Link Forecasting on Temporal Knowledge Graphs
   code https://github.com/liu-yushan/TLogic
6. Long Horizon Forecasting With Temporal Point Processes
   code https://github.com/pratham16cse/DualTPP
7. Time-Series Event Prediction with Evolutionary State Graph
   code https://github.com/VachelHU/EvoNet
8. Dynamic Gaussian Mixture Based Deep Generative Model for Robust Forecasting on Sparse Multivariate Time Series
   code https://github.com/thuwuyinjun/DGM2
9. Spatio-Temporal Recurrent Networks for Event-Based Optical Flow Estimation
   code https://github.com/ruizhao26/STE-FlowNet

I want to ask how the stgnn model works

Hi,

Thank you for the great work.

This is more of a suggestion than an issue.

Different papers use different weather datasets. So far I have found two: one is from USA weather stations with one-hour aggregation and the other is from Max Plank institute with 10 mins aggregation. Moreover, it looks like different authors are using the Max Plank dataset differently. For example, some of them are using 2 years of data while others are only using one.

I thought about bringing this up in case you want to bring this level of detail. If you want I could help :)

Best.

This is a very cool project!

Hi @lixus7

Thanks for your excellent repo!
Would you mind considering our paper published on ICDE 2023 about diffusion-based imputation models for spatiotemporal time series?
Title: PriSTI: A Conditional Diffusion Framework for Spatiotemporal Imputation
Paper Link: https://arxiv.org/abs/2302.09746
Code Link: https://github.com/LMZZML/PriSTI

例如方法ST-SSL中所说的BJTaxi数据集就是TaxiBJ