Nikronic / Data-Mining-IUST

This repository contains the assignments of Data Mining course taught by Dr. Minaei at IUST

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Data-Mining-IUST

This repository contains the assignments of Data Mining course taught by Dr. Minaei at IUST

Contents

You can find explanation in each folder. In this section, there is just summary of each work folder:

Anomaly Detection in Stream using RRCF

In this work, I have prepared a report of paper RRCF by Amazon which has been used to detect anomaly in stream data. This folder contains the report of the paper and presentation slides.

Data Mining in Dota 2

In this work, I have introduced Dota 2 game, as a complex multi-objective, multi-agent problem and introduced different usage of regression, clustering, etc on high-level manner as a possible solution to issues like balancing the game, proper matchmaking, etc. This folder only contains a report file.

Fast Decision Trees

In this work, Fast Decision Trees has been introduced as the replace for the conventional decision tree to compensate slow interaction for large trees.

HW01

This work contains information about using different topics such as:

  • Different data mining techniques that is being used in search engines
  • Different distances
  • Dimensionality reduction
  • Undersampling
  • Data cleansing.

HW02

In this work, following topics has been covered:

  • Different clustering methods
  • SMC and Jaccard distance
  • Implementation of DBSCAN from scratch

MiniHW01

In this work, a majority of prominent metrics in data mining field have been compared to each other for different use cases. Also their benefits and downsides have been discussed too.

Rule vs. Tree

The differences between rule learning approaches and tree based approaches have been discussed in this work.

Project

In this work, DeepPath paper has been discussed thoroughly that is a deep learning based approach to the Graph Knowledge Reasoning task. I have introduced a few ideas to improve the original ideas based on some other papers which you can find the ideas on the report and presentation files. Also, a fork of original repository has been made to apply new changes(WIP).

Acknowledgment

I have used many contents from different websites which I have not referenced! I have bookmarked all those references and I will update the slides or at least provide the links at each folder.

Course Content

Slides can be shared upon request. Please send me an email (available on nikronic.github.io or if you could not find it, contact me by creating an issue. Videos are also available but I highly suggest you to use another resource.

Citation

  • Add citation
  • create zenodo page

About

This repository contains the assignments of Data Mining course taught by Dr. Minaei at IUST

License:Apache License 2.0


Languages

Language:Jupyter Notebook 89.5%Language:HTML 5.1%Language:TeX 2.8%Language:Python 2.6%Language:Shell 0.0%