Pinned Repositories
autonomous-car
bostonhousepricing
data-flows
Pocket data flows orchestrated using Prefect
Distributed-Slice-mobility-attack-detection
Inter-slice mobility in 5G networks allows mobility of user sessions from one network slice to another. A novel targeted attack against network slices of 5G networks by exploiting the user equipment-initiated inter-slice mobility will be mentioned here. We name this attack as distributed slice mobility (DSM) attack. The performance and economic damage caused by the DSM attack is higher than the denial-of-service and Yo-Yo attacks. We will deep learning model to detect this kind of attack efficiently
MLOps_best_practices
Movie-analysis-by-Apache-Spark
Data cleaning, preprocessing, and analyze on a million movies using Apache Pyspark
Optimizing-Resoruce-Scaling-for-Network-Slicing-through-Attention-based-Forecasting
This project will build a deep learning-based forecasting model which leveraging an attention mechanism to forecast the future usage of Virtual network function instances' resources in network slices and then scale up and down the number of VNF instances based on the predicted usage
Product-review-classification
In this project, I build a LSTM-based sentiment classification model to classify custormer's behaviour buying clothes and jewerly in the Amazon website based on their reviews of ordered products leaving on the website. The model's output is a 3-class output which are postivie, negative and neutral. This project uses a pretrained word2vec model which is Google Word2Vec model to embed sentences into word embedding vectors. Then the LSTM model will use these embedding vectors to train the model.
Seoul_air_quality_forecast
Two-phase-Deep-learning-based-EDoS-Detection-System
Cloud computing is currently considered the most cost-effective platform for offering business and consumer IT services over the Internet. However, it is prone to new vulnerabilities. A new type of attack, called an economic denial of sustainability (EDoS) attack, exploits the pay-per-use model to scale up the resource usage over time to the extent that the cloud user has to pay for the unexpected usage charge. In this project, we proposed a two-phase deep learning-based detection system to detect EDoS attack. The first phase called the prediod detector will detect where there is an attack in a period of 5s and then trigger the second phase detector if there is an attack in that 5-second period. The second detector called the flow detector will detect abnormal flows in the abnormal period detected by the first detector.
harrychien1311's Repositories
harrychien1311/Optimizing-Resoruce-Scaling-for-Network-Slicing-through-Attention-based-Forecasting
This project will build a deep learning-based forecasting model which leveraging an attention mechanism to forecast the future usage of Virtual network function instances' resources in network slices and then scale up and down the number of VNF instances based on the predicted usage
harrychien1311/Distributed-Slice-mobility-attack-detection
Inter-slice mobility in 5G networks allows mobility of user sessions from one network slice to another. A novel targeted attack against network slices of 5G networks by exploiting the user equipment-initiated inter-slice mobility will be mentioned here. We name this attack as distributed slice mobility (DSM) attack. The performance and economic damage caused by the DSM attack is higher than the denial-of-service and Yo-Yo attacks. We will deep learning model to detect this kind of attack efficiently
harrychien1311/autonomous-car
harrychien1311/bostonhousepricing
harrychien1311/data-flows
Pocket data flows orchestrated using Prefect
harrychien1311/MLOps_best_practices
harrychien1311/Movie-analysis-by-Apache-Spark
Data cleaning, preprocessing, and analyze on a million movies using Apache Pyspark
harrychien1311/Product-review-classification
In this project, I build a LSTM-based sentiment classification model to classify custormer's behaviour buying clothes and jewerly in the Amazon website based on their reviews of ordered products leaving on the website. The model's output is a 3-class output which are postivie, negative and neutral. This project uses a pretrained word2vec model which is Google Word2Vec model to embed sentences into word embedding vectors. Then the LSTM model will use these embedding vectors to train the model.
harrychien1311/Seoul_air_quality_forecast
harrychien1311/Two-phase-Deep-learning-based-EDoS-Detection-System
Cloud computing is currently considered the most cost-effective platform for offering business and consumer IT services over the Internet. However, it is prone to new vulnerabilities. A new type of attack, called an economic denial of sustainability (EDoS) attack, exploits the pay-per-use model to scale up the resource usage over time to the extent that the cloud user has to pay for the unexpected usage charge. In this project, we proposed a two-phase deep learning-based detection system to detect EDoS attack. The first phase called the prediod detector will detect where there is an attack in a period of 5s and then trigger the second phase detector if there is an attack in that 5-second period. The second detector called the flow detector will detect abnormal flows in the abnormal period detected by the first detector.
harrychien1311/DDoS-Detection-using-deep-learning
In this project, I used ANN (Artificial Neural network) to detect DDoS attack
harrychien1311/harrychien1311
Config files for my GitHub profile.
harrychien1311/llama3-chatbot
This soucre code is the inference pipeline of LLama3 which can run in Linux locallay
harrychien1311/Markdown-cheatsheet
harrychien1311/picar1
harrychien1311/prefect-docker
Demo on how to use Prefect with Docker
harrychien1311/prefect-google-trends
A data workflow fetching and creating google trends reports based on keywords. This workflow is automated by Prefect
harrychien1311/Protein-Sequence-Analysis
In this repo, I analyzed the protein sequence extracted by using 2 tools Javelin and Skeleton-productions of Bionsight. Based on the analyzed result I can conclude which tool give a better performance in finding protein sequence.
harrychien1311/starter-hugo-academic
harrychien1311/ViLT-for-visual-question-answering
This source code is forked from ViLT repo of the ICML 2021 (long talk) paper: "ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision". I modified the data processing part to process my data for the 2024 VizWiz Grand Challenge