How to Parse Resume/CV using POI

[Mujahid Ateeb]

Hi All,
 How to extract the candidate name, education, experience, project details form resume using poi. Currently i'm using keyword search and regex patteren matching but it displays all data if one keyword matches.

import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

import javax.annotation.Resource;

import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFHeader;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;
import org.apache.poi.xwpf.usermodel.XWPFTable;
import org.apache.solr.client.solrj.SolrQuery;
import org.apache.solr.client.solrj.SolrServerException;
import org.apache.solr.client.solrj.impl.HttpSolrClient;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;

import com.rtp.portal.dao.ParsingKeywordsDao;
import com.rtp.portal.solr.model.ParseKeywords;


@Service("parseKeywordsService")
public class ParseKeywordsService {
	
	@Autowired
	HttpSolrClient solr;
	
	@Resource
	ParsingKeywordsDao parsingKeywordsDao;
	
	//public enum categories { PASSPORT, PAN, EXPECTEDSALARY, LANGUAGES, DOB, MARITALSTATUS }  
	
	public void parseResume(String resume)
	{
		storeKeywordsInSolr();
		QueryResponse response;
		List<String> passport = null;
		List<String> pan = null;
		List<String> dofb = null;
		List<String> expectedSalry = null;
		List<String> languages = null;
		List<String> maritalStatus = null;
		List<String> education = null;
		List<String> skills = null;

		try {
			response = solr.query(new SolrQuery("*:*"));
			List<ParseKeywords> foundDocuments = response.getBeans(ParseKeywords.class);
			for(ParseKeywords keys:foundDocuments)
			{
				passport=keys.getPassport();
				pan=keys.getPan();
				dofb=keys.getDob();
				expectedSalry=keys.getExpectedSalry();
				languages=keys.getLanguages();
				maritalStatus=keys.getMaritalStatus();
				education=keys.getEducation();
				skills=keys.getSkills();
			}
		} catch (SolrServerException | IOException e) {
			// TODO Auto-generated catch block
			e.printStackTrace();
		}
		
		try {
            File file = new File(resume);
            FileInputStream fis = new FileInputStream(file.getAbsolutePath());

            @SuppressWarnings("resource")
			XWPFDocument document = new XWPFDocument(fis);

            List<XWPFParagraph> paragraphs = document.getParagraphs();
            List<XWPFHeader> headers=document.getHeaderList();
            List<XWPFTable> tables = document.getTables();
            //String regex = "^[_A-Za-z0-9-\\+]+(\\.[_A-Za-z0-9-]+)*@[A-Za-z0-9-]+(\\.[A-Za-z0-9]+)*(\\.[A-Za-z]{2,})$";
            //Pattern p = Pattern.compile(regex);
            for(int i=0;i<headers.size();i++)
            {
            	//System.out.println(headers.get(i));
            }
          
            for(int i=0;i<tables.size();i++)
            {
            	String tableData[]=tables.get(i).getText().split("\n");
            	for(int j=0;j<tableData.length;j++){
            	for(String pp:passport){
                	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(pp))
                	if(isWordExist(tableData[j],pp))
            		{
                		System.out.println(tableData[j] + " tpass"+ j);
                		break;}
                	}
                	for(String panc:pan){
                    	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(panc))
                		if(isWordExist(tableData[j],panc))
                		{
                    		System.out.println(tableData[j] + " tpan"+ j);
                    		break;}
                    }
                	for(String dob:dofb){
                    	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(dob))
                		if(isWordExist(tableData[j],dob))
                		{
                    		System.out.println(tableData[j] + " tdob"+ j);
                    		break;}
                    }
                	for(String expected:expectedSalry){
                    	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(expected))
                		if(isWordExist(tableData[j],expected))
                		{
                    		System.out.println(tableData[j] + " texpt"+ j);
                    		break;}
                    }
                	for(String language:languages){
                    	if(tableData[j].trim().replaceAll("\\s{2,}", " ").toLowerCase().contains(language))
                		//if(isWordExist(tables.get(i).getText().toLowerCase(),language))
                		{
                    		System.out.println(tableData[j] + " tlang"+ j);
                    		break;}
                    }
                	for(String ms:maritalStatus){
                    	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").toLowerCase().contains(language))
                		if(isWordExist(tableData[j].toLowerCase().replaceAll("\\s{2,}", " ").trim(),ms))
                		{
                    		System.out.println(tables.get(i).getText() + " tms"+ j);
                    		break;}
                    }
                	
                	for(String edu:education){
                    	if(tableData[j].trim().replaceAll("\\s{2,}", " ").contains(edu))
                		//if(isWordExist(tables.get(i).getText().toLowerCase().replaceAll("\\s{2,}", " ").trim(),edu))
                		{
                    		Pattern regex = Pattern.compile("[0-9]%");
              	           Matcher matcher = regex.matcher(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " "));
              	           
              	           if(matcher.find()){
                     		System.out.println(paragraphs.get(i).getText() + " tedu"+ i);
              	           }
                    		break;}
                    }
                	for(String skill:skills){
                    	if(tableData[j].trim().replaceAll("\\s{2,}", " ").contains(skill))
                		//if(isWordExist(tables.get(i).getText().toLowerCase().replaceAll("\\s{2,}", " ").trim(),edu))
                		{
                    		System.out.println(tableData[j] + " tskills"+ j);
                    		break;}
                    }
            	}
            }
            for(int i=0;i<paragraphs.size();i++)
            {
            	
            	for(String pp:passport){
            	//if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(pp))
            	if(isWordExist(paragraphs.get(i).getText(),pp))
            	{
            		System.out.println(paragraphs.get(i).getText() + " pass"+ i);
            		break;}
            	}
            	for(String panc:pan){
            		Pattern pattern = Pattern.compile(panc);
            		//if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(" "+panc+" "))
            		if(isWordExist(paragraphs.get(i).getText(),panc))
            		{
                		System.out.println(paragraphs.get(i).getText() + " pan"+ i);
                		break;
                	}
                }
            	for(String dob:dofb){
                	//if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(dob))
                	if(isWordExist(paragraphs.get(i).getText(),dob))
            		{
                		System.out.println(paragraphs.get(i).getText()+ " dob"+ i);
                		break;}
                }
            	for(String expected:expectedSalry){
                	//if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(expected))
            		if(isWordExist(paragraphs.get(i).getText(), expected))
            		{
                		System.out.println(paragraphs.get(i).getText()+ " exp"+ i);
                		break;}
                }
            	for(String language:languages){
                	if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").toLowerCase().contains(language))
                	//if(isWordExist(paragraphs.get(i).getText(),language))
            		{
                		System.out.println(paragraphs.get(i).getText()+ " lang"+ i);
                		break;}
                }
            	for(String ms:maritalStatus){
                	//if(tables.get(i).getText().trim().replaceAll("\\s{2,}", " ").toLowerCase().contains(language))
            		if(isWordExist(paragraphs.get(i).getText().toLowerCase().replaceAll("\\s{2,}", " ").trim(),ms))
            		{
                		System.out.println(paragraphs.get(i).getText() + " ms"+ i);
                		break;}
                }
            	for(String edu:education){
                	if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(edu+" "))
            		//if(isWordExist(paragraphs.get(i).getText().replaceAll("\\s{2,}", " ").trim(),edu))
            		{
                		Pattern regex = Pattern.compile("[0-9]%");
         	           Matcher matcher = regex.matcher(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " "));
         	           
         	           if(matcher.find()){
                		System.out.println(paragraphs.get(i).getText() + " edu"+ i);
         	           }
                		break;}
                }
            	for(String skill:skills){
                	if(paragraphs.get(i).getText().trim().replaceAll("\\s{2,}", " ").contains(skill+" "))
            		//if(isWordExist(paragraphs.get(i).getText().replaceAll("\\s{2,}", " ").trim(),edu))
            		{
                		System.out.println(paragraphs.get(i).getText() + " skills"+ i);
                		break;}
                }
            }
            
            fis.close();
        } catch (Exception e) {
            e.printStackTrace();
        }
	}
	
	private static boolean isWordExist(String sentence, String word){
	    boolean ans = Boolean.FALSE;        
	    ArrayList<String> wordList = null;

	    try {

	        if(sentence != null && word != null){
	            wordList = new ArrayList<String>(Arrays.asList(sentence.split("[^a-zA-z]+")));              
	            if(wordList.contains(word)){
	                ans = Boolean.TRUE;
	            }
	        }
	    } catch (Exception e) {
	        e.printStackTrace();
	        // TODO: handle exception
	    }
	    return ans;
	}
	
	public void storeKeywordsInSolr()
	{
		List<Object[]> list=parsingKeywordsDao.getKeywords();
		List<String> passport = new ArrayList<String>();
		List<String> pan = new ArrayList<String>();
		List<String> dofb = new ArrayList<String>();
		List<String> expectedSalry = new ArrayList<String>();
		List<String> languages = new ArrayList<String>();
		List<String> maritalStatus = new ArrayList<String>();
		List<String> education = new ArrayList<String>();
		List<String> skills = new ArrayList<String>();
		
		for(Object[] obj:list)
		{
			if(((String)obj[0]).toUpperCase().equals("PASSPORT"))
				passport.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("DOB"))
				dofb.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("EXPECTEDSALRY"))
				expectedSalry.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("LANGUAGES"))
				languages.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("PAN"))
				pan.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("MARITALSTATUS"))
				maritalStatus.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("EDUCATION"))
				education.add((String)obj[1]);
			if(((String)obj[0]).toUpperCase().equals("SKILLS"))
				skills.add((String)obj[1]);
		}
		ParseKeywords keys =new ParseKeywords();
		keys.setDob(dofb);
		keys.setExpectedSalry(expectedSalry);
		keys.setId("0");
		keys.setLanguages(languages);
		keys.setMaritalStatus(maritalStatus);
		keys.setPan(pan);
		keys.setEducation(education);
		keys.setPassport(passport);
		keys.setSkills(skills);
		
		try {
			solr.addBean(keys);
			solr.commit();
		} catch (IOException | SolrServerException e) {
			// TODO Auto-generated catch block
			e.printStackTrace();
		}
	}
}

[Tim Moores]

That's a lot of code. Please given an example input, and tell us what this code does with it, and what should happen instead.

My guess would be that CVs are formatted in too many different ways to be amenable to be processed via regexps.

[Campbell Ritchie]

Presumably you would have a web form which the CV is entered with, so you can parse each of the fields. Again it would probably not be a format amenable to parsing with regexes: what will you do if candidate A says, “BSc in Computer Sciences,” and candidate B says, “Computer Sciences BSc”?
Why are you parsing names in the first place? That is not a feature you want to assess before interview.

And welcome to the Ranch

[Mujahid Ateeb]

Tim Moores wrote:That's a lot of code. Please given an example input, and tell us what this code does with it, and what should happen instead.

My guess would be that CVs are formatted in too many different ways to be amenable to be processed via regexps.

Hi, Can you tell me any method or any algorithm to extract education,project details and experience from companies.

Here I'm using regex and String methods.
If I'm going to search MS education it prints MS education as well as skills like MS Office.

[Tim Moores]

Unless the input is regularized in some fixed format, that's impossible to achieve in general. The formats of CVs that people come up with are much too varied for such a strictly described approach. As Campbell mentioned, the same concept can be conveyed in multiple ways, to say nothing of misspellings.