Document Ranking Algorithm

//then, we look at relative occurrences
//herein lies the most disgusting code i've ever written
for(int i = 0; i < toSearch.length; i++){

    //find the line we need in results
    //j is irrelevant, just trying to find the correct results line
    for(int j = 0; j < results.size(); j++){
        if(results.get(j).split(":")[0].equals(toSearch[i].toLowerCase())){
            //we got it
            //look through and populate the array

            String curSplit[] = results.get(j).split(":");
            //for each relevent document, add its array spot
            //k is current array spot
            for(int k = 1; k < curSplit.length; k++){

                String curEntry = curSplit[k];
                String curDoc = curEntry.split("->")[0];

                if(debug){
                    System.out.println(curEntry.split("->")[0] + " is relevant at array spot " + k);
                }

                //find relevant document's name
                for(int l = 0; l < docNames.size(); l++){
                    if(docNames.get(l).equals(curDoc)){
                        Integer curNum = docs.get(docNames.get(l));

                        if(debug){
                            System.out.println("curNum = " + curNum);
                        }

                        curNum += k;
                        if(debug){
                            System.out.println(docNames.get(l) + "'s current relevancy: " + curNum);
                        }

                        docs.put(docNames.get(l), curNum);
                        break;
                    }
                }
            }
        }
    }
}