The WikAPIdia Java framework provides easy and efficient access to multi-lingual Wikipedia data.

###Main features

• Support for all Wikipedia language editions and comparisons between multiple languages.
• Tools that download and organize Wikipedia datasets published by the Wikimedia foundation.
• Tools that load downloaded Wikipedia datasets into databases for analysis.
• Tools that identify multi-lingual concepts and the pages in each language edition that link to those concepts.
• Semantic-relatedness algorithms that measure the relationship between two concepts such as "racecar" and "engine."
• Single-machine parallelization (i.e. multi-threading support) for all computationally intensive features.

###System Requirements

• Maven (required)
• Bash (required)
• A clone of this repository
• Hardware varies depending on the languages you want to import:
  • Simple English (175K articles) requires a few GB and 10 minutes of processing on a four core laptop.
  • Full English (4M articles) requires 200GB and 6 hours of processing on an eight core server.

###Importing data

• Clone this repository git-clone https://github.com/shilad/wikAPIdia.git
• Download and process the dataset:

	cd wikAPIdia
	cd wikAPIdia-parent
	./scripts/runpipeline.sh all -l simple

(Note: be sure to run code from the folders indicated above)

The last command downloads, installs, and analyzes the latest database files for the Simple English langauge edition of Wikipedia.

You can customize WikAPIdia's importing procedure, but the run-pipeline-sh script should be a good start. For example, you can specify different language editions by changing the -l parameters. To analyze English and French you could run:

./scripts/run-pipeline all -l en,fr

(beware that this is a lot of data!).

###An example program Once you have imported data (above), your are ready to write programs that analyze Wikipedia! Here's a simple example you can find in the Cookbook:

// Prepare the environment; set the root to the current directory (".").
Env env = new EnvBuilder()
        .setBaseDir(".")
        .build();

// Get the configurator that creates components and a phraze analyzer from it 
Configurator configurator = env.getConfigurator();
PhraseAnalyzer pa = configurator.get(PhraseAnalyzer.class);

// get the most common phrases in simple  
Language simple = Language.getByLangCode("simple");   // simple english 
LinkedHashMap<LocalPage, Float> resolution = pa.resolveLocal(simple, "apple", 5);
        
// show the closest pages
System.out.println("resolution of apple");
if (resolution == null) { 
    System.out.println("\tno resolution !");
} else {
    for (LocalPage p : resolution.keySet()) {
        System.out.println("\t" + p + ": " + resolution.get(p));
    }       
} 

When you run this program, you'll see output:

resolution of apple
	LocalPage{nameSpace=ARTICLE, title=Apple, localId=39, language=Simple English}: 0.070175424
	LocalPage{nameSpace=ARTICLE, title=Apple juice, localId=19351, language=Simple English}: 0.043859642
	LocalPage{nameSpace=ARTICLE, title=Apple Macintosh, localId=517, language=Simple English}: 0.043859642
	LocalPage{nameSpace=ARTICLE, title=Apple Inc., localId=7111, language=Simple English}: 0.043859642
	LocalPage{nameSpace=ARTICLE, title=Apple A4, localId=251288, language=Simple English}: 0.043859642

###A tour of the example Let's walk through this program to explain each piece. First, we create an ```Env``, a WikAPIdia environment that provides access to the components we need:

Env env = new EnvBuilder()
        .setBaseDir(".")
        .build();

The EnvBuilder provides utility methods to set the languages you want to support, the maximum number of threads available to your program, etc. There are more advanced ways of configuring WikAPIdia - both programatically and through configuration files - described in the Configuration section of this page.

The Env provides access to a Configurator - essentially a Factory for creating WikAPIdia components. We get the Page Resolution component next:

Configurator configurator = env.getConfigurator();
PhraseAnalyzer pa = configurator.get(PhraseAnalyzer.class);

A key feature of WikAPIdia is that it supports multiple implementations of the same component. For example, the default PhraseAnalayzer uses the Lucene search engine. We could have explicitly requested the lucene implementation of the PhraseAnalyzer:

PhraseAnalyzer pa = configurator.get(PhraseAnalyzer.class, "lucene");

If we instead wanted to use a phrase analyzer that resolves phrases to pages by looking at "intra-wiki" links, we could have used:

PhraseAnalyzer pa = configurator.get(PhraseAnalyzer.class, "anchortext");

And received the results:

resolution of apple
	LocalPage{nameSpace=ARTICLE, title=Apple, localId=39, language=Simple English}: 0.55263156
	LocalPage{nameSpace=ARTICLE, title=Apple Inc., localId=7111, language=Simple English}: 0.30526316
	LocalPage{nameSpace=ARTICLE, title=Apple Records, localId=47698, language=Simple English}: 0.12631579
	LocalPage{nameSpace=ARTICLE, title=App Store (iOS), localId=216566, language=Simple English}: 0.010526316
	LocalPage{nameSpace=ARTICLE, title=Apple Corps, localId=48013, language=Simple English}: 0.005263158

###Main components The WikAPIdia Configurator offers a set of components that you can use as building blocks in your application. To get one of these components, use the Configurator.get() method. TODO: List and one-sentence description of most important components

###Configuration The behavior of WikAPIdia can be customized through configuration files or code. The default WikAPIdia configuration is determined by the main reference.conf. The configuration is backed by Typesafe config and uses the HOCON format. To override the configuration settings create your own configuration file containing the changes to reference.conf and pass it to the EnvBuilder.

For example, suppose we wantd to set the root directory, maxThreads to 8, and the phrase analyzer to the anchortext-based analyzer We could create a file called myapp.conf containing:

maxThreads : 8
baseDir : /my/path/to/wikAPIdia
phrases.analyzer.default : anchortext

We would then tell the EnvBuilder to use the new config to override the default settings:

Env env = new EnvBuilder()
        .setConfigFile("./path/to/myapp.conf")
        .build();

We could also make these changes directly in Java, without the config file:

Env env = new EnvBuilder()
        .setMaxThreads(8)
        .setBaseDir('/my/path/to/wikAPIdia')
        .setProperty('phrases.analyzer.default', 'anchortext')
        .build();