// prepare bi-gram tokenizer factory
Map<String, String> args = new HashMap<String, String>();
args.put(AbstractAnalysisFactory.LUCENE_MATCH_VERSION_PARAM, "4.4");
args.put("minGramSize","2");
args.put("maxGramSize","2");
TokenizerFactory tf = new NGramTokenizerFactory(args);
// (ab)->(bc)->(cd)->[ef][fg][gh]
List<String> rules = new ArrayList<String>();
rules.add( "abcd=>efgh" );
synMap = new SlowSynonymMap( true );