String stemming

Question

hi everyone,
i have a program that searches a file for a string. i need to be able to implement functionality to this progam so that it can find like strings. for example when i search for run, i want runner and running to also be displayed as well as run, i have thought about just storing an array of possible stems and searching for them aswell as the original string, however this wont suffice as i need to be able stem strings that aren't always familiar to me. i have included my search method code, i search a file by first reading it into a string and then spliting this string into other strings; title, abstract, body, keyword and index. i am then searching each of these strings for a target string using the indexof(). does anyone have any idea how i will be able to implement string stemming functionality to this?
public void search() {
        String paf = pathText.getText();
        path = new File(paf);
        target = searchText.getText();
        if (path.isFile()){
            searchfile(paf);
        if (path.isDirectory()){
            //folderlen = path.length();
            processdir(path);
        MouseListener ml = new MouseAdapter() {
           public void mousePressed(MouseEvent e) {
        if ((list.getSelectedIndex() != -1) && (e.getClickCount() == 2))  {
            String selection = (String)list.getSelectedValue();
            display(selection);
                list.addMouseListener(ml);
    public void processdir(File path) {
        int i;
        File[] theFiles = path.listFiles();
        for (i = 0; i < theFiles.length; i++) {
            if(theFiles.isDirectory()){
//theFiles = path.listFiles();
processdir((theFiles[i]));
if(theFiles[i].isFile()) {
String pop= theFiles[i].getAbsolutePath();
//System.out.print(pop+"\n");
searchfile(pop);
//list.setListData(files);
public void searchfile(String pah){
pqs = new File(pah);
String co = readFile(pqs);
ArrayList text = documentSpliter(co);
int rank = 0;
String pan = pqs.getName();
String st= pqs.getPath();
if (title.indexOf(target)>=0) {
rank+=10;
files.add("\nthe string was found in the title");
boolean intit = true;
if (abs.indexOf(target)>= 0) {
rank+=5;
files.add("\nthe string was found in the abstract");
boolean inabs = true;
if (body.indexOf(target)>= 0) {
rank+=2;
files.add("\nthe string was found in the body");
boolean inbod = true;
if (keyword.indexOf(target)>= 0) {
rank+=3;
files.add("\nthe string was found in the keyword");
boolean inkey = true;
if (index.indexOf(target)>= 0) {
rank+=2;
files.add("\nthe string was found in the index");
boolean inind = true;
if (rank==0){
files.add("\nword not found in file");
boolean eof = true;
if((rank!=0)) {
files.add(pan+" "+"rank score:"+rank);
files.add("\nthe document "+pan+" has rating:"+rank+"\n\n");
files.add("\n");
files.add(st+"\n");
list.setListData(files);

Answer 1

I think the only likely approach is to have a table of common word extensions ("-es" "-ing" "-ed" "-ly" and so on). You might extend that to actual rules (e.g. for plural formation) and exceptions.