SPMF is a popular and highly efficient data mining software written in Java, specialized in pattern mining. It provides over 300 algorithms for various tasks such as:
- Frequent itemset mining
- Association rule mining
- Sequential pattern mining
- Sequential rule mining
- High-utility itemset mining
- Episode mining
- Graph mining
- Time series analysis
- and more
SPMF offers a graphical user interface (GUI), a command-line interface, and a server for alternatively running data mining algorithms through REST queries from a Python or Web client. SPMF can also be integrated in Python, R and other languages through wrappers and its CLI, or used as a Java library in Java projects. SPMF is lightweight, actively developed and has no external dependencies.
The latest release is SPMF version v2.65, released on February 18, 2026.
The official website of SPMF with full documentation, tutorials, and other resources is: http://philippe-fournier-viger.com/spmf/
- Quickstart
- Documentation
- Datasets
- Architecture
- Related Resources
- Contributing
- License
- How to Cite SPMF
- Authors
There are five ways to use SPMF, depending on your needs:
The most simple way to use SPMF is through its integrated GUI. To run the GUI:
- Download the files
spmf.jarandtest_files.zipto your computer and make sure that Java is installed on your computer. - Uncompress the file
test_files.zipon your desktop. It will create a folder containing some example data files that you can use with the algorithms. - Launch the GUI of SPMF by double-clicking on the file
spmf.jar. If it does not work and you are using Windows, right-click onspmf.jarand select "open with..." and then select "Java Platform". If this option is not there, perhaps that Java is not installed on your computer, or that the PATH environment variable does not include your Java installation. - If the previous step succeeds, the graphical interface of SPMF will open.
- Then, from the user interface, you can select input files, choose an algorithm from more than 300 algorithms, sets its parameters, and run the algorithms. For example, let's say that you want to run the CM-SPAM algorithm. In the documentation, it is the CM-SPAM example. To run that example:
- Click on the combo box beside "Choose an algorithm" and select "CM-SPAM".
- Click on "Choose input file" and select
contextPrefixSpan.txtfrom thetest_filesfolder. - Click on "Choose output file", select a location (e.g., Desktop), enter
result.txt, and click "OK". - Set the parameter minsup (%) to
0.5(as in the example). - Other parameters are optional and can be ignored for this example.
- Click on "Run algorithm".
- A new window will open showing the results.
- The results correspond to the patterns discovered by CM-SPAM (see the CM-SPAM example of the documentation for interpretation). That’s all. To run another algorithm, follow the same steps.
The second to use SPMF is through its command line interface (CLI) from the console. To run the SPMF using the CLI:
- Download the files
spmf.jarandtest_files.zipto your computer and make sure that Java is installed on your computer. - Uncompress the file
test_files.zipon your desktop. It will create a folder containing some example data files that you can use with the algorithms. - To run an algorithm, go to the documentation of SPMF find the example corresponding to the algorithm that you want to run. For example, let's say that you want to run the PrefixSpan algorithm. It is this example in the documentation.
- Open the command prompt (if you are using Windows) or the terminal (if you are using Linux). Then, type the command specified in the example. For example, for PrefixSpan, the command is:
java -jar spmf.jar run PrefixSpan contextPrefixSpan.txt output.txt 50%This command means to run the algorithm named "PrefixSpan" to use the input file named "contextPrefixSpan.txt" to set the output file for the results as "output.txt" to set the parameter of the algorithm (minsup) to 50 %. - After executing this command, the file output.txt will be created. It will contain the result.
That's all. If you want to run another algorithm, then follow the same steps.
The third way to use SPMF is by integrating it into Java projects. For this, you can download spmf.jar and include it in the classpath of your Java project, and then call the classes from SPMF from your Java program. Or alternatively, you can download spmf.zip or clone the project to obtain SPMF source code.
To get a good grasp of how the source code of SPMF is organized, you may read the developers guide.
The fourth way to use SPMF is to call it from language developed using other programming languages using wrappers for Python, R, C#, etc. via community-provided wrappers that invoke the command-line interface of SPMF. A list of wrappers is here.
The fifth way to use SPMF is through the SPMF-Server, a lightweight HTTP server that wraps the SPMF library and exposes all algorithms as a REST API. This lets any language or tool submit mining jobs over HTTP and retrieve results without needing a local Java integration. This can be useful to run SPMF on a remote machine and query it from a client, from the browser or integrate it into a web application or microservice. Currently, the SPMF server can be used with the SPMF Server Python CLI and GUI clients or the SPMF Server Web client. For more details about how to install and run the SPMF-Server, please see the SPMF-Server project.
The main documentation of SPMF and other resources can be found on the SPMF website.
- List of algorithms: https://philippe-fournier-viger.com/spmf/index.php?link=algorithms.php
- Main documentation (with examples for each algorithm): https://philippe-fournier-viger.com/spmf/index.php?link=documentation.php
- Release notes and download download page
- FAQ: https://philippe-fournier-viger.com/spmf/index.php?link=FAQ.php
To run experiments with SPMF, multiple datasets are provided on the SPMF website, in SPMF format: https://philippe-fournier-viger.com/spmf/index.php?link=datasets.php
A general overview of the architecture of SPMF is provided below.
To use SPMF, a user can choose to use the Graphical interface, Command line interface or the SPMF-server. The user interacts with any of these interfaces to run algorithms which are managed by a module called the Agorithm Manager. There are mainly three types of algorithms, which are (1) data pre-processing algorithms, (2) data mining algorithms, and (3) algorithms to either visualize data or patterns found in the data. The Algorithm Manager has the list of all available algorithms, and a description of each algorithm. The description of an algorithm indicates how many parameters it has, what are the data types of parameters, what is the algorithm name, etc. The input and output of algorithms are generally text files. A few different formats are supported, explained in the documentation of SPMF.
The source code is organized in several packages. The main packages are:
ca.pfv.spmf/
│
├── algorithms/
│ ├── associationrules/ → Association rule mining algorithms
│ ├── classifiers/ → Classification algorithms
│ ├── clustering/ → Clustering algorithms
│ ├── episodes/ → Episode mining algorithms
│ ├── frequentpatterns/ → Itemset mining algorithms
│ ├── graph_mining/ → Graph mining algorithms
│ ├── sequenceprediction/ → Sequence prediction algorithms
│ ├── sequential_rules/ → Sequential rule mining algorithms
│ ├── sequentialpatterns/ → Sequential pattern mining algorithms
│ ├── sort/ → Sorting algorithms
│ └── timeseries/ → Time series mining & analysis algorithms
│
├── algorithmmanager/
│ ├── Algorithm Manager → Central registry for algorithms
│ └── descriptions/ → Metadata (input/output types, authors, etc.)
│
├── datastructures/ → Specialized data structures (e.g., triangular matrix)
├── gui/ → Graphical User Interface (MainWindow.java)
│ └── Main.java → Command-line entry point
│
├── input/ → Input file readers (transactions, sequences, etc.)
│
├── patterns/ → Pattern representations (itemsets, rules, etc.)
│
├── test/ → Example usage of algorithms (developer samples, not unit tests)
│
└── tools/ → Utilities (generators, converters, statistics, etc.)
- The SPMF website
- SPMF-Server — REST API server for SPMF
- SPMF Server Python client — Python CLI and GUI clients for SPMF-Server
- SPMF Server Web client. A Web client (HTML+JS+CSS) for the SPMF-Server
- The Pattern Mining Course — A free online course covering pattern mining algorithms and their implementation
- More Pattern Mining Videos on the @philfv YouTube channel
- The Data Blog — Blog from the founder of SPMF
- Other Resources — Books, tutorials, links to other projects, etc.
If you would like to contribute improvements, please contact the SPMF founder at philfv AT qq DOT com. In particular, if you want to contribute new algorithms not yet implemented in SPMF, you are very welcome to get in touch.
See the contributors page for a full list of people who have contributed to the project.
The source code and files in this project are licensed under the GNU General Public License v3.0 (GPLv3). The GPL license grants four freedoms:
- Run the program for any purpose
- Access the source code
- Modify the source code
- Redistribute modified versions
Restrictions: If you redistribute the software (or derivative works), you must:
- Provide access to the source code
- License derivative works under the same GPLv3 license
- Include prominent notices stating that you modified the code, along with the modification date
For full details about the license and its requirements, see the GPLv3 license.
If you use SPMF in your research, please cite one of the following papers:
-
Fournier-Viger, P., et al. (2012). *SPMF: A Java Open-Source Pattern Mining Library. Journal of Machine Learning Research (JMLR).
-
Fournier-Viger, P., Lin, C.W., Gomariz, A., Gueniche, T., Soltani, A., Deng, Z., Lam, H. T. (2016). The SPMF Open-Source Data Mining Library Version 2. In Proceedings of the 19th European Conference on Principles of Data Mining and Knowledge Discovery (PKDD 2016), Part III, Springer LNCS 9853, pp. 36–40.
For a full list of citations, see the citations page. Citing SPMF helps support the project — thank you! 🙏
Project Leaders:
- Prof. Philippe Fournier-Viger (Founder), https://www.philippe-fournier-viger.com/ (e-mail: philfv AT qq DOT com)
- Prof. Jerry Chun-Wei Lin
- Prof. Wei Song — North China University of Technology, Beijing, China
- Prof. Vincent S. Tseng — National Chiao Tung University, Taiwan
- Prof. Ji Zhang — University of Southern Queensland, Australia, https://staffprofile.unisq.edu.au/Profile/Ji-Zhang
Contributors: A full list of all contributors can be found at: https://www.philippe-fournier-viger.com/spmf/index.php?link=contributors.php