Java API for Auto regression (AR), ARIMA, Time Series Analysis [closed] - java

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 6 years ago.
Improve this question
I am looking for either Opensource or Free Java API for Time Series Analysis using AR, ARIMA etc. I need this api for DDOS Attack analysis.
I googled around and found 2 solutions but both are not completely solving the problem:
1) This same question was asked earlier in stackoverflow and a solution was posted regarding SuanSu Api but this API is not free
2) Apache Math Library, but this API provides other forms of Regression like Simple, OLS, GLS etc but not Auto Regression.
I checked for Options in Machine Learning apis like Mahout but not luck yet. Please suggest an appropiate API

I spent my 4th year Computing project on implementing time series forecasting for Java heap usage prediction using ARIMA, Holt Winters etc, so I might be in a good position to advise you on this.
Your best option by far is using the R language, you can call on the forecasting libraries provided by R, through Java by using the JRI library found here. R is well documented, free and open source. You can even run R on a server and then make calls to it via command line using Rserve, which then returns forecasts over HTTP but JRI is the local equivalent if memory serves me correctly.
If you have any questions, let me know.

Have a look at spark-timeseries. The source code is mostly Scala, but it's relatively simple to use the library from within Java. If you're in a place where you are doing time series analysis on the JVM, then you should consider learning Spark/Scala anyways.
The library is young as of this writing and has room for improvement and growth, but as of version 0.3 it implements AR, ARIMA, simple exponential smoothing (EWMA), and Holt-Winters smoothing. It's areas for improvement are a better automatic ARIMA algorithm, support for seasonal ARIMA, and state space modeling, but it's already very useful.

Related

Are there any open source Java analysis libraries for data/control dependence? [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 3 years ago.
Improve this question
I am looking for a library/API that has some out of the box data and control dependence analyses for Java programs. This would be a static analysis on the control flow graph (CFG) of the given program to compute data dependences and control dependences. I would like to avoid reimplementing these techniques/algorithms if they have already been done. An inter-procedural analysis would be great, but I could work with a library that does an intra-procedural one as well.
Any suggestions would be greatly appreciated.
Edit: One thing I have found is jChord, but I haven't been able to determine yet if it actually has any of the out of the box functionality that I am looking for or if I would have to implement it myself.
Another possibility is the joeq library which seems to at least have the infrastructure necessary for doing this, but the documentation (or lack thereof) is making it difficult for me to tell what it is actually capable of.
I am sure Eclipse does plenty of data flow analysis underneath the hood, but I haven't seen anything yet that is public facing. Anyone know of the Eclipse API having stuff like this?
Try http://www.sable.mcgill.ca/soot/
OP says he is interested in non-open source systems too.
Our DMS Software Reengineering Toolkit with its Java Front End can parse Java source code in all dialects 1.4-1.7, producing full ASTs, build symbol tables, compute types of expressions, and determine control and dataflows within methods, including explicity control dependence as requested by OP.
Usually folks that are interested in advanced analyses have something other than the raw analysis in mind. DMS is an ecosystem of program analysis and transformation tools, that can be used to leverage such analyses into diagnostics about the existing code in terms of source location (drawn directly from the ASTs) or source code (prettyprinted from a subtree of interest), or to generate new code fragments (by assembling ASTs and prettyprinting them) or finally by actually changing the original code (by modifying the ASTs using procedural modifications or better yet, source-to-source transformation, and prettyprinting the modified AST).

Examples of Object-Oriented Projects Help Procedural Programmers [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Please help me identify some small to medium sized open source projects that embody object oriented design (preferably in C++ or Java). I would like to use these projects to demonstrate how real world problems (as opposed to contrived text book examples) can be solved with an object oriented design. I want to be able to present a plausible explanation of why certain things were chosen to be objects and how they all work together to solve a problem.
Google Chromium (C++): windows, tabs, plugins etc. are all classes.
The Unreal Tournament Public Source Code (432 Headers) contains the declarations of the Unreal engine class library written in C++. I found it to be a rich example of a large object-oriented program. It taught me a lot about how to modularize and object-orient my code. It also demonstrates many tactics for getting a handle on a large code base.
Also, because all you can read are header files, you'll have a fun (and educational) time trying to figure out how the whole thing comes together. (I actually ended up writing my own x86 disassembler so I could cheat and read some of the definitions!)
On the same note, the Doom 3 SDK contains a large chunk of the Doom/Quake engine written in very readable C++.
Just about any large project designed in Java is object-oriented, almost by definition. You can take a look at Apache Hadoop as a large-scale, open-source, objected oriented project written in Java. Another is Apache Ant.
Eclipse would be a good example on the Java side: the plugin architecture is all object oriented.
I asked the same thing to my OO mentor. He pointed me to the JUnit sources, with the recommendation to see how it evolved version by version. This would show you how Kent Beck writes Java code.
Another example on this vein would be the sources of Fit by Ward Cunningham.

Visualizing Data in Java [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 8 months ago.
Improve this question
Is there a good library on the web to visualize big data in Java? Perhaps a library for statistical outputs? I know the programming language R is used to visualize statistical data in R, and I have seen a solution to connect Java and R, but it would be better if I can have a pure Java solution.
might want to check out http://processing.org/
I've had success with visualizing graphs with a commercial tool called yFiles. For more general purpose statistical data visualization, you could try Mondrian, which is GPL licensed. These both are Java libraries.
Have to mention Piccolo2D -- using it very frequently. It does not provide the ready to use data structures as Prefuse does, but it surely scales and provides a great degree of freedom in how one would want to visualise the data.
JFreeChart is a good Option. I have tried it, and is easy to set up with Eclipse too. BIRT is also nice, but if you are using it for reporting that is.
Prefuse is a pretty nice library for Java, and even lets you publish animations in Flash using a layer called Flare:
prefuse visualization toolkit
prefuse visualization gallery
On the R side, you only need to browse the gallery to see what it's capable of:
R Graphics Gallery
There's a library for Clojure (not quite Java but still on the JVM) called Incanter.
It is built on Colt and JFreeChart. It may be possible that you can just use Colt directly, but I wouldn't know.
The Weka project might be able to help you. It's an open source library of data mining algorithms written in Java. If they don't have the visualization themselves, it might be in their "related projects" section.
http://www.cs.waikato.ac.nz/ml/weka/

Java to C cross compilation [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 6 years ago.
Improve this question
Does anyone know of a good Java to C cross compiler?
Do they work well?
This is very similar to this question, and answers may be helpful to you: Compiler to translate Java to C.
Summary: There are tools for this (Toba, GCJ, etc), but you may run into problems with not all of the Java libraries being ported. In the end, tools will probably only do PART of the work, and you'll have to hand-code some of the rest.
A good first step is to convert your Java code to only use standard libraries available in Java 1.4. In fact, you'll probably want to wean as much as possible off of anything not in java.lang.* or java.util.* packages in order to simplify the porting procedure.
Depending on the size of your codebase, it may actually be easier to rewrite the bulk directly rather than relying on tools. Java and C have a lot of syntax similarity, but the mismatch between C's straight procedural code, and Java's object oriented features could cause problems. Automated tools may generate virtually unmaintainable C code when trying to work around this, and there's always the possibility for subtle bugs.
2016 update: Don't do this, not now, not ever. The options that used to provide this have not been maintained (GCJ, for example), and it's arguably easier to find a developer fluent in java than C. Also, Java performance has continued to improve, and baseline implementations tend to have similar performance. Optimized C is still faster, but the edge gets smaller and smaller with every JRE version.
Can you explain why you want to port your Java code to c?
If it's for performance you likely won't see much of an improvement. Java is a garbage collected language and currently there isn't an algorithm that can insert memory allocation and deallocation calls efficiently. There have been many researchers trying to solve this problem and they have some interesting solutions but I have not seen a good commercial product that can scale to large programs yet. You can look at the conference proceeding for previous ISMM conferences for more information.
If you want to speed your code up I suggest that you use a profiler and find the hot methods. Try and optimize the hot methods and if that is still not enough try and use JNI.

Is there a good NumPy clone for Jython? [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
I'm a relatively new convert to Python. I've written some code to grab/graph data from various sources to automate some weekly reports and forecasts. I've been intrigued by the Jython concept, and would like to port some Python code that I've written to Jython. In order to do this quickly, I need a NumPy clone for Jython (or Java). Is there anything like this out there?
I can't find anything that's a clone of numpy, but there's a long list of Java numerics packages here - these should all be usable from Jython. Which one meets your requirements depends on what you're doing with numpy, I guess.
Wilberforce is essentially corrrect.
However, I suggest looking at the Apache Commons Math library -- that would be a better choice for a replacement Java numerics package than any of those listed in wilberforce's answer.
Incanter, a Clojure scientific/statistical computing library, uses the Parallel Colt Java libraries with great success: http://incanter.org/. One route may be to start using the PColt classes in Jython, and slowly build up Python-esque bindings for it, as Incanter provides? (Let me know if you have interest in this.)
There is a build called JNumeric available on sourceforge:
The sourceforge version has not had a release in a long time, but it seems like an updated version for Jython 2.51 is also available (have not tried it myself):
http://bitbucket.org/zornslemon/jnumeric-ra/downloads/

Categories