Engineers and scientists use MATLAB® to organize, clean, and analyze complex data sets from diverse fields such as climatology, predictive maintenance, medical research, and finance. MATLAB provides:
- Datatypes and preprocessing capabilities designed for engineering and scientific data
- Interactive and highly customizable data visualizations
- Apps and Live Editor tasks that helps with interactive data cleaning, preparation, and code generation
- Thousands of prebuilt functions for statistical analysis, machine learning, and signal processing
- Extensive and professionally written documentation
- Accelerated performance with simple code changes and additional hardware
- Expanded analysis to big data without big code changes
- Automatic packaging of analysis into freely distributable software components or embeddable source code without manually recoding algorithms
- Sharable reports automatically generated from your analysis
Using MATLAB for Data Analysis
Organize and Explore Data
Organize your data with datatypes designed for tabular, time-series, categorical, and text data. Use the MATLAB language to write programs based on thousands of algorithms from a wide variety of domains. Interactively customize visualizations, then automatically generate the MATLAB code to reproduce them with new data.
Analyze and Clean Data with Less Code
MATLAB Live Editor tasks and apps allow you to interactively perform iterative tasks such as cleaning data, training machine learning models or labeling data. These tasks and apps then generate the MATLAB code needed to programmatically reproduce the work you did interactively.
Use a prebuilt family of functions for identifying and cleaning sensor drift, signal outliers, missing data, and noise. Combine separate data sets by joining tables and synchronizing time-series data. Live Editor Tasks let you interactively solve these problems within your live script and generate the code for you. The Data Cleaner app helps to identify data problems and iteratively configure and apply multiple cleaning methods to clean time series data.
Easily Scale Up Your Analysis
parfor loops and multiprocessor hardware to accelerate parallel analysis with almost no code changes. Create
gpuarrays to take advantage of GPU acceleration for appropriate algorithms. Process out-of-memory data sets using tall arrays, which overload hundreds of functions throughout the data analysis workflow to operate on out-of-memory data.
Share Your Results
Package your analysis in freely shareable software components such as executables, C/C++ libraries, .NET assemblies, Java® libraries, and Python® packages. Automatically translate your MATLAB code to C and C++ code for deployment to embedded targets. Document your work using MATLAB Live Editor and export your results to reports in PDF, Microsoft® Word, Latex, and HTML.