KO
|
EN
gitlite — search
Search
#python
#javascript
#css
#css3
#html
#linux
#android
#reactjs
#sql
#minecraft
#discord
#github
SAFE
★ 35
Open GitHub ↗
No description available.
Download README (.md)
Explore Similar Repositories
PytorchSR
:
Pytorch based phoneme recognition (TIMIT phoneme classification)
gin-stats
:
Gin's middleware for request stats
yancheng-sales
:
天池-印象盐城-汽车销量预测大赛
GraphqlDockerProxy
:
A generic Graphql API for Docker and Kubernetes
NetAPP
:
本项目（项目地址：https://github.com/MartinWuQing/NetAPP）用的是MUI做的前台 后台用的是ASP.NET Core 技术。（项目地址： https://github.com/MartinWuQing/mhqNetAPP），是做服务器后台以及后台管理界面开发的项目，
// repository documentation
Was this content helpful?
★ 0
(0 ratings)
Select Rating:
★
★
★
★
★
Submit Feedback
Recent Feedback
×
Download README
Do you want to download the
README.md
file for
SAFE
?
Download (.md)
# SAFE: A Neural Survival Analysis Model for Fraud Early Detection In this paper, we propose a survival analysis based fraud early detection model, SAFE, that maps dynamic user activities to survival probabilities that are guaranteed to be monotonically decreasing along time. ## Running Environment The main packages you need to install ``` 1. python 2.7 2. tensorflow 1.3.0 ``` ## DateSet For experiments, we evaluate **SAFE** on two real-world datasets: twitter and wiki which have been attached in [twitter/](https://github.com/PanpanZheng/SAFE/tree/master/twitter) and [wiki/](https://github.com/PanpanZheng/SAFE/tree/master/wiki), respectively. ## Model Evaluation The command lines for SAFE and baselines go as follow * **SAFE** ``` python framework/safe.py $1 ``` * **M-LSTM** ``` python framework/base_rnn.py $1 ``` * **CPH & SVM** ``` python framework/safe_baselines.py $1 ``` **where** *$1* refers to datasets on which the model runs, and it can be assigned as 'twitter' or 'wiki'. * **Weibull & other distributions** ``` python framework/safe_distr.py $1 $2 ``` **where** *$1* refers to the corresponding distributions and it can be assigned as 'exp' (exponential), 'ray' (Rayleigh) and 'poi' (poisson); *$2* denotes the datasets, 'twitter' or 'wiki'. ## Authors * **Panpan Zheng, Shuhan Yuan and Xintao Wu** - [personal website](https://sites.uark.edu/pzheng/) - [google scholar](https://scholar.google.com/citations?user=f2OLKMYAAAAJ&hl=en) ## Citation I am very glad that you could visit this github and check my research work. If it benefits your work, please cite the paper in Arxiv https://arxiv.org/abs/1809.04683v1 . ## Acknowledgments This work was going on underlying the guide of prof. [Xintao Wu](http://csce.uark.edu/~xintaowu/) and Dr. [Shuhan Yuan](https://sites.uark.edu/sy005/). Appreciate it greatly for every labmate in [**SAIL lab**](https://sail.uark.edu/)
