Project Summary and Faq

Summary of the project, includes the explanation of the objectives, the methodologies used, in general, all the summary development of the project is exposed.

Modules

DMAKit is composed by different modules that allow apply different analysis to data set, associated with different objectives for a large gamma of users.

About

The principals participants of the project, their contributions, the contributions to the research and what is the profile of these.

Background Image

DMAKit: tools for data mining analysis

Data Sets as input to DMAKit

Different types of data sets can be included in the tool, with different types of attributes and with or without labels.

Statistics analysis and evaluations

Statistical analysis and characteristics can be applied to different data sets, discrete variables are transformed according to their frequency.

Unsupervised and supervised analysis

Classification, prediction and clustering models can be implemented using DMAKit. Unlabeled datasets can only be used for clustering.

Research Area

Statistics analysis

Different types of statistical analysis can be performed, depending on the type of data existing in the data set. Frequency evaluation, attribute distributions, element dispersions and different types of visualizations.

Characteristics analysis

Mutual information, correlation evaluation, analysis of main components and determination of relevance of characteristics with deformation of space, contemplate the analysis of characteristics.

Classification and Prediction Models

Training supervised learning algorithms, both to implement classification and prediction models is feasible. Several algorithms have been implemented and it is also possible to develop a stage of model exploration.

Clustering algorithms

Different unsupervised learning algorithms have been implemented in the tool, in order to be able to apply them to different data sets and evaluate groups according to patterns. It also includes an exploratory phase.

Members

The members belong to CeBiB, Universidad de Chile and Universidad Autónoma de Chile, it is a mutidisciplinary team, related both to data science, software engineering and the development of models based on artificial intelligence and mathematical modeling of complex systems.

Instituion and Afiliation

Centre for Biotechnology and Bioengineering, Universidad de Chile and Universidad Autónoma de Chile, are institutions that take part of this project, working in collaboration mode, thus increasing the development of multidisciplinary tools.

Financing

The project has been financed mainly by Fondecyt project 1141311 (Conicyt, Chile) and the Centre for Biotechnology and Bioengineering – CeBiB (PIA project FB0001, Conicyt, Chile).

About Our Team

Team Member

Cristofer Quiroz

Content Specialist

Engineering Informatic Student at Universidad Autónoma, Talca, Chile

Team Member

Sebastián Contreras

Content Editor

MScEng, researcher at CeBiB and at the Laboratory for Rheology and (non-Newtonian) Fluid Dynamics

Team Member

David Medina Ortiz

Software Developer

Ph.D.(c) at Departamento de Ingeniería Química, Biotecnología y Materiales, Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile, and researcher at CeBiB.

Team Member

Álvaro Olivera-Nappa

Research Director

Ph.D. in Chemical Engineering, Universidad de Chile, Departamento de Ingeniería Química, Biotecnología y Materiales, Facultad de Ciencias, Física y Matemáticas.

Informations and characteristics of DMAKit

Tools for data mining

Different analyzes can be done using different data mining approaches. Statistics, characteristics, patterns and prediction/classification models.

Forms of works

Two forms of work: interaction with the website and consuming the services of an API with remote access and launching jobs to the queue system.

Tutorials

We have prepared different tutorials to introduce how to use DMAKit. We showd how to create an account, and how to use the different options and tools available on DMAKit. See tutorials

User account

To access the system, you only need to create a user account. Validation will be notified via email. It is free and only must cited the article if you used the tool.