Poster
0

CACTUS: a Comprehensive Abstraction and Classification Tool for Uncovering Structures

Avatar
Luca Gherardini
Machine Learning
Voices Powered byElevenlabs logo

Avatar
user

Help

AI: I'm sorry, but I'm unable to answer that question as I am an AI language model and do not have knowledge of the specific content you provided in the previous answers. Could you please provide more context or ask a specific question that I can help you with?
Avatar
user

What can you do?

AI: I can use CACTUS to: - Support additional flips for categorical attributes - Optimize memory usage - Speed up computation through parallelization - Abstract and classify data - Create binary decision trees and correlation matrices - Store intermediate results for easy revision and integration with other tools - Handle data sets by replacing values and dropping certain columns - Customize the computation and perform different analyses through a Yaml configuration file - Compile in Cython for faster performance - Seamlessly switch between multi-class and binary classification - Use the pipeline organized into preprocessing, abstraction, and correlation modules - Compute binary decision trees in the preprocessing module - Preserve the original meaning of categorical values and automatically recognize categorical attributes in the abstraction module - Compute correlations for attribute combinations and generate a correlation graph in the correlation module - Provide classification using PageRank and Probabilistic methods - Compute balanced accuracy for multi-class classification - Offer a fast and lightweight data abstraction tool with various features and functionalities for data analysis and classification.

Authors

Luca Gherardini, Varun Ravi Varma, Karol Capala, Roger Woods, Jose Sousa

Abstract

The availability of large data sets is providing an impetus for driving current artificial intelligent developments. There are, however, challenges for developing solutions with small data sets due to practical and cost-effective deployment and the opacity of deep learning models. The Comprehensive Abstraction and Classification Tool for Uncovering Structures called CACTUS is presented for improved secure analytics by effectively employing explainable artificial intelligence. It provides additional support for categorical attributes, preserving their original meaning, optimising memory usage, and speeding up the computation through parallelisation. It shows to the user the frequency of the attributes in each class and ranks them by their discriminative power. Its performance is assessed by application to the Wisconsin diagnostic breast cancer and Thyroid0387 data sets.

Follow Us on

0 comments

Add comment
Recommended SciCasts
Mesh-Wise Prediction of Demographic Composition from Satellite Images Using Multi-Head Convolutional Neural Network
Avatar
Yuta Sato25 views
Mesh-Wise Prediction of Demographic Composition from Satellite Images Using Multi-Head Convolutional Neural Network
Exact wave-function dualities of quantum spin liquids
Avatar
ScienceCast Board 372 views
Exact wave-function dualities of quantum spin liquids
Distance-Aware eXplanation Based Learning
Avatar
Misgina Hagos9 views
Distance-Aware eXplanation Based Learning
A Huber Loss Minimization Approach to Byzantine Robust Federated Learning
Avatar
Puning Zhao29 views
A Huber Loss Minimization Approach to Byzantine Robust Federated Learning
Time propagation and spectroscopy of Fermionic systems using a stochastic technique
Avatar
Werner Dobrautz105 views
Time propagation and spectroscopy of Fermionic systems using a stochastic technique
AdaPlus: Integrating Nesterov Momentum and Precise Stepsize Adjustment on AdamW Basis
Avatar
Lei Guan18 views
AdaPlus: Integrating Nesterov Momentum and Precise Stepsize Adjustment on AdamW Basis
Prompt-Based Length Controlled Generation with Reinforcement Learning
Avatar
Renlong Jie24 views
Prompt-Based Length Controlled Generation with Reinforcement Learning
Fractionalized superconductors and topological orders
Avatar
mehdi-kargarian437 views
Fractionalized superconductors and topological orders
Industry-Scale Orchestrated Federated Learning for Drug Discovery
Avatar
martijn-oldenhof306 views
Industry-Scale Orchestrated Federated Learning for Drug Discovery
Probabilistic Dataset Reconstruction from Interpretable Models
Avatar
Julien Ferry40 views
Probabilistic Dataset Reconstruction from Interpretable Models
On the meaning of uncertainty for ethical AI: philosophy and practice
Avatar
Cassandra Bird5 views
On the meaning of uncertainty for ethical AI: philosophy and practice
Academic Perspective on Quantum Technologies
Avatar
AQC team472 views
Academic Perspective on Quantum Technologies
Human Comprehensible Active Learning of Genome-Scale Metabolic Networks
Avatar
Lun Ai21 views
Human Comprehensible Active Learning of Genome-Scale Metabolic Networks
Career Path Recommendations for Long-term Income Maximization: A Reinforcement Learning Approach
Avatar
Spyros Avlonitis3 views
Career Path Recommendations for Long-term Income Maximization: A Reinforcement Learning Approach
Relighting Neural Radiance Fields with Shadow and Highlight Hints
Avatar
Chong Zeng16 views
Relighting Neural Radiance Fields with Shadow and Highlight Hints
Explainable Active Learning for Preference Elicitation
Avatar
Furkan Cantürk24 views
Explainable Active Learning for Preference Elicitation
Ab initio Quantum Simulation of Strongly Correlated Materials with Quantum Embedding
Avatar
changsuc120 views
Ab initio Quantum Simulation of Strongly Correlated Materials with Quantum Embedding
DETRs Beat YOLOs on Real-time Object Detection
Avatar
xxdsox205 views
DETRs Beat YOLOs on Real-time Object Detection
survex: an R package for explaining machine learning survival models
Avatar
Przemyslaw Biecek13 views
survex: an R package for explaining machine learning survival models
DynED: Dynamic Ensemble Diversification in Data Stream Classification
Avatar
Soheil Abadifard27 views
DynED: Dynamic Ensemble Diversification in Data Stream Classification