Machine learning is one of the most hot technologies, and major companies are actively recruiting related programming talents to fill the vacancies written by machine learning and depth learning code. It is true that according to the relevant recruitment statistics, the Python language has exceeded Java to become an employer's most urgent machine learning programming skill. But in fact, Java still plays an irreplaceable role in project development, and many popular machine learning frameworks are written by Java. In view of the previous reference information about Python, as fresh as Java, today we recommend five industry top Java machine learning libraries.
WEKA
Address: http://www.cs.waikato.ac.nz/ml/weka/index.html
There is no doubt that Weka is currently preferred Java machine learning library.
It developed by the New Zealand University of Waikato. It is named from a New Zealand's unique bird - New Zealand, the English name of New Zealand, the English name of New Zealand is Weka. According to the official website description, Weka absorbs many commonly used machine learning algorithms, and is completely based on Java environment, open source, free, free graphical interface, suitable for data mining, data analysis, and predictive modeling and other application scenarios.
EIBE FRANK from Computer Science from Waikato University: "The biggest advantage of Weka is classification, so applications that require automatic data classification can benefit. But it also supports data pretreatment, clustering, and association rules. Time series prediction, feature selection, and abnormal detection and other scenes. "
Developers can directly process target data sets through the Weka software, and also support users in their own code to be called, regard Weka as a flexible component. More intimate is that Waikato University also offers many free Weka-based data mining and machine learning video tutorials, interested friends can click on Weka official website.
Massive Online Analysis (MOA)
Address: http://moa.cms.waikato.ac.nz/
And WEKA, MOA's naming is also from a unique wing big bird in New Zealand - New Zealand fear bird (currently extinct). MOA is also based on Java environment, open source, free, MOA can work with Weka when facing complex issues.
MOA has made special optimization in terms of operational efficiency and memory usage, by providing an easy-to-extended underlying structure, which can be diverted data stream analysis settings and a series of internal implementation of machine learning algorithms, MOA provides a very very data stream analysis. Excellent reference framework, therefore the application in real-time data stream mining fields is very wide. The internal implementation of the machine learning algorithm includes: classification, regression, clustering, isolated point detection, concept drift detection, and recommended system. In addition, MOA also provides a variety of assessment tools, as well as active community discussions, blogs and other supporting resources.
DeePlearning4j
Address: https://deeplearning4j.org/
DeePlearning 4J (Deep Learning for Java) is an open source distributed depth study project in Java and Scala environments, which is headquartered in business intelligence and enterprise software company SKYMIND, headquartered in San Francisco, and has got Tencent's investment. As it is named, DeePlearning4j's run requires support for Java virtual machine JVM.
Last year, the JAXENTER community nominated Deeplearning4j as one of the most innovative contributors of the Java ecosystem.
DeePlearning4j aims to provide a flexible DIY machine learning tool for Java, Scala and Clojure programmers working under the Hadoop framework. The team said in the official website that they hope to use some machine learning algorithms to bring business to intelligent data. It is also to achieve this ideal, benefiting more users, thus choosing a better Java environment to achieve these algorithms. At present, DeePlearning4j is widely used in pattern identification, time series detection, and voice, text-of-the-art emotional identification, including Google, Facebook, and Microsoft and other giant companies are its users.
Mallet
Address: http://mallet.cs.umass.edu/
MALLET is primarily developed by Professor Andrew McCallum from the University of Massachusetts and its student, is an open source machine learning toolkit based on Java environment. Mainly used in statistical natural language processing, document classification, clustering, topic modeling, information extraction and other text class analysis scene.
Mallet implements many powerful tools, including advanced tools for document classification, tools for sequence tags, and tools for theme modeling. MALLET also supports various types of algorithms, including simple Bayes, decision trees, and maximum entropy. In addition, Mallet also provides many routines, including particulars, deletions, and converts text to vector representations.
Elki
Address: https: //lki-project.github.io/
The full name of ELKI is: Environment For Developing Kdd-Applications Supported by Index-Structures, the KDD application development environment supported by the index structure, where KDD refers to Knowledge Discovery IN Database, ie knowledge discovery.
Elki is also a well-known Java-based data mining software. ELKI's focus is algorithm research, emphasis on cluster analysis, database index, and isolation point detection. ELKI can independently analyze data mining algorithms and data management tasks, which is unique in other data mining frames such as WETA and RAPIDMINERs. In addition, ELKI also supports various data types and file formats, as well as various similarity methods.
Elki is designed for researchers and student, providing a large number of flexible algorithm parameters for simple and fair algorithms. At present, ELKI has been widely used in related fields of various data science, including whales echo positioning, aerospace operation, shared bicycle distribution, and traffic forecasting.
Search for "Love Bo.com" to pay attention, daily update development board, intelligent hardware, open source hardware, activities, etc., you can make you master. Recommended attention!
[WeChat scanning picture can be directly paid]
Related Reading:
Sad reminder! The Wintel Alliance, which is a decades of the calculation industry, began to fall.
Our other product:
