PDF Processing and Resume Detection Freelancer -- 2
Budget: $50 – $100 CAD
We are seeking a highly skilled and experienced C# developer to write a program that can process PDF files and detect resumes. The program must be able to extract text from the PDFs using iTextSharp and use a set of criteria to determine if a PDF is a resume or not. The program must be able to process at least 300-500 resumes per second on an average computer (iCore 5, 16GB RAM).
Project Description:
1. Extract text from PDFs using iTextSharp or similar libraries
2. Implement a set of criteria to detect if a PDF is a resume or not, including the number of pages, existence of certain key phrases as headers (such as "Work Experiences" and "Education"), existence of skills (such as "C#/Java/etc"), existence of action words (such as "performed, designed, developed"), existence of date ranges and company names, existence of address, and most importantly existence of phone numbers and email or LinkedIn profile (a resume without contact information is not a resume for our purposes)
3. Ensure that the program processes at least 300 resumes per second on an average computer (iCore 5, 16GB RAM)
4. Write clean, efficient, and well-documented code
5. Test the program to ensure that it is working as intended
** Measurement of success (Milestone release condition):
1. Must have 92%+ Accuracy
- Accuracy = (True Positives + True Negative) divided by (Number of Resumes Processed)
2. Must process at least 300 resumes per second
Project Description:
1. Extract text from PDFs using iTextSharp or similar libraries
2. Implement a set of criteria to detect if a PDF is a resume or not, including the number of pages, existence of certain key phrases as headers (such as "Work Experiences" and "Education"), existence of skills (such as "C#/Java/etc"), existence of action words (such as "performed, designed, developed"), existence of date ranges and company names, existence of address, and most importantly existence of phone numbers and email or LinkedIn profile (a resume without contact information is not a resume for our purposes)
3. Ensure that the program processes at least 300 resumes per second on an average computer (iCore 5, 16GB RAM)
4. Write clean, efficient, and well-documented code
5. Test the program to ensure that it is working as intended
** Measurement of success (Milestone release condition):
1. Must have 92%+ Accuracy
- Accuracy = (True Positives + True Negative) divided by (Number of Resumes Processed)
2. Must process at least 300 resumes per second