Modern Resume Analyser For Students And Organisations

e-ISSN:2395-0056

Volume: 09 Issue: 10 | Oct 2022 www.irjet.net p-ISSN:2395-0072

Modern Resume Analyser For Students And Organisations

Vansh Nawander1, Shrenita Elma2, Andhoju Karthikeya3, Manuka Koushik Yadav4, Sai Karthik Kotala5, M. D. N. Akash6

1,2,3,4,5 Student, Vardhaman College of Engineering, Hyderabad, India 6 Business Mentor, Rejolt Edtech Private Limited, Hyderabad, India ***

Abstract - : The Resume Analyser is a streamlit-based web application that analyzes an uploaded resume in the format of pdf and extracts information. It returns a resume score based on the presence of key elements such as a declaration, a list of hobbies, and qualifications. This analyzer reduces the need for people to manually check their resumes and gives them confidence in their resumes. This analyzer works using natural language processing and textminingtodeduce informationfromtheresume.

Key Words: Natural Language Processing(NLP), Text mining, streamlit.

1. INTRODUCTION

In today's world, everyone possesses skills but many of themcanbarelygetahighlypaidjob.Inspiteofhavingthe skills, people are failing. The lack of quality projection of the resume during the interviews is the main reason. So for a fresher in order to present a perfect resume that projects the skills precisely, our resume analyser helps in building the perfectresume byanalyzingcoreconceptsof ML.Thereisonlyoneopportunityforyoutomakeagood impression on the recruiter and receive an interview invitation. Even a personal, cordial consultation with a well-knownHRManagerisnoassurancethatyourresume will be accepted by the ATS. By running your resume through a review tool, you may identify 21 typical problems, learn about the formatting, and get helpful advice on how to stand out to recruiters. The Smart Resume Analyzer System is a text mining application that analyses resumes received by an organization using keyword matching algorithms. The keywords from the personalized dictionary will be mapped against the terms in the resume with the aid of the keyword matching algorithm. The necessary data is extracted and placed in the database once the keywords have been matched. The complete information is sorted using many criteria, including experience, education, age, and others. The system then schedulesinterviews based on the generated sorted list. Through an automated email that the system generates,applicantsareinformedoftheinterview.

2. LITERATURE SURVEY

Intoday'sworld,thedevelopmentoftechnologiessuchas the Internet is revolutionizing the hiring process. There areover50,000onlinejobsitesthatencourageapplicants to submit their resumes on their websites. It is very unlikely that you will be selected from thousands of applications.This smart resume analyzer may be very useful in growing your possibility of being on the candidate list. It offers pointers concerning your resume writingskills.

The RDF(Resource Description Framework) ontology of CV was presented by Uldis Bojars and John G. Breslin. It uses the RDF data model to model the CV. Information about resumes is described using a wide range of classes and characteristics in resume RDF. In order to properly explainthematerial,UldisBojarshasexpandedFOAFwith resume details. In 2002 and 2003, Turney and Littmann took into account the semantic relevance of other words he identified as a paradigm and took into account the semantic orientation or evaluation characteristics of wordsfromavastcorpusof100billionwordsprovidinga schematoguess.

UjjalMarjitet.al[1]suggestedanotherwaytorecoverCV information using the idea of linked data. This allows the Internettosharedata with differentsources anddiscover differenttypesofinformation.

An online Chinese resume parser was developed by Zhi Xiang Jing et. al [2] extracted data from the CV using a rule-basedstatisticalalgorithm.

Zhang Chuang et. al [3] developed the largest CV analysis system by working on block analysis of CV documents based on pattern matching and multi-level information identification.

Elik et. al [4]intended thefollowing systemtoconvert CV totheontologicalstructure.

A model that simplifies Turkish and Turkish analysis English CV. Di Wu et. al [5] Successful Better extract

International Research Journal of Engineering and Technology (IRJET)

e-ISSN:2395-0056

Volume: 09 Issue: 10 | Oct 2022 www.irjet.net p-ISSN:2395-0072

information from your resume through the idea of the ontologyusingWordNet Calculationofsimilarity.

Top resume analyzed using techniques such as natural language processing. With the help of natural language processing,onlythetextdataisextractedfromtheCVand the strength of the applicant profile is displayed as a percentage. Additional attributes such as B. Percentage of candidateskillsasa resultofa resumereview,depending on the candidate's education, qualifications, courses, and work experience. However, there is no provision on the website that it applies only to job seekers. Posting a particular job may give recruiters within your organization the opportunity to provide a ranking of candidates that match your immediate job-related skills. Therefore, there are many other such web applications, mostofwhichprovidesimilarfunctionality.

To improve the hiring process, Mayuri Verma proposed the Cluster based Ranking Index (CBR), which ranks resumestolocatethebestcandidates.[6]Inthepast,very few efforts were made to make a Recommender System. Then again numerous researches have been found on Recommender SystemResnick et.al developed [7] the first recommendationsystem.

Otaibi et al. explored the use of recruitment services, described the actions to take during each organization's recruitment process, and how organizations can benefit from electronic recruitment portals., what candidate criteria might lead to the selection, and a variety of other essentialrecruitingapproaches.R.Jananiet.alpresenteda text mining research using Natural Language Processing(NLP),clustering.Theefficiencyowasincreased using text mining tools in the extraction point.D. Cerrick and others have published a paper entitled Toward an Ontology-Based Information Extraction System that Matches work with CV.In this,plain text resume is converted into ontology form by Ontology Knowledge Base(OKB). This system calculates the completion rate according to work experience, education, etc. Marinovsky et. al To create an employment proposal using an Expectation-Maximization(EM)algorithmthattakes both thecandidate'sresumeandtheemployer'sjobdescription into consideration. Golecetal has proposed for determiningtherelevanceofacandidate'sjob description based on fuzzy method. Text analysis also explains the fast-growing field Text Mining, defined as statistical machine learning how to translate unstructured raw materials .You can transform information into structured information for further classification, classification, sufficient training and training. Excellent functional information [8]. Technology .Pattern recognition, information extraction, data Mining and parsing are

involved in text mining, called Natural Language Processing (NLP). Even if there are various several other websites for resume analysis, our website is unique in its own way. In our we have two interfaces one is admin interface and the other is user interface.In the admin interface you view the candidates who have used our website in an excel. Recruiters can put the users resume on the candidate list based on their resumescore and the skills required for a particular position. In this way, it helps the recruiters. On the other hand,you have user interface where you can upload your resume and get the resumescoreandtherecommendationsrequiredtoboost yourresumescore.

3. PROPOSED APPROACH

Working of Resume Analyser uses a considerable number ofpythonmoduleswhichmakesthehandlingandworking oftheapplicationeasier.Someofthenotablemodulesare streamlit, Resumeparser, pandas, pdfminer3, matplotlib, pysql. Roles of different components used are: since streamlit is used to create web applications for data scienceandmachinelearninginashorttime,anditisalso more structured and focused more on simplicity, so streamlit is used in smart resume analyzer. By using Resumeparser,itconvertsanunstructuredformofresume dataintoastructuredformat.Itanalysesresumedataand extracts it into machine-readable output such as XML, JSON.Resumeparser automaticallystores,organizes, and analyses resume data to find the best candidate. Pandas module is used in manipulation of files like csv or other types of files. And it is also used in manipulating the dataframes, series, etc. Matplotlib library used to manipulate the data using the visualization tools such as pie charts in weighing the skills of the candidate/user. Instead of using MYSQL or some other databases, for smoother access of data PySQL is being used. PySQL is a purely python independant SQL database that is free of other dependancies enabling us to implement the application easily. In order to analyse the or extract the exact text/data from the uploaded pdf file from the user thismoduleisdedicated.

Web applications continue to evolve at an incredible rate, andthearchitecturearoundwebapplicationsisbecoming more and more complex. Most web applications rely on client-server architecture, where the client gives information and the server stores and retrieves the information. Most web applications available on the Internet are written in programming languages such as HTML, CSS for designing and animations, and Javascript used to create the front-end interface(client-side program). For scripting web applications, server-side programming is written using programming languages

International Research Journal of Engineering and Technology (IRJET)

e-ISSN:2395-0056

Volume: 09 Issue: 10 | Oct 2022 www.irjet.net p-ISSN:2395-0072

likePython,Java,PHP,andRuby,etc.PythonandJavahave commonly used languages for server-side programming. For storing the data from the web applications local databases like MySQL, Oracle, PostgreSQL, etc. are used [15]. A resume analyzer is a streamlit-based web application that analyzes an uploaded resume in the format of pdf and extracts information. This analysis is done with the help of python libraries that include pdfminer,nltk, spacy, pandas, NumPy, etc. The web application has two sections, mainly the normal user and admin panel. The user can upload his/her resume in the user section with a limit of pdf file upto 200 megabytes. Upon upload, the site user will be able to see his/her resumeandtheanalysisoftheresumewhichincludesthe user'scurrentskills,recommendedskills,resumescoreof theuser,andtheleveloftheuseraspertheanalysis.This analysis helps the user to check for improvements in his/her resume which could help him to get better opportunities.theresumedatawhichisuploadednowgets stored in the local mysql database which can be accessed by the admin only through username and password. The adminsectionwhichunlocksthroughpasswordhasvisual analysisofthetotalresume;thisanalysisisrepresentedin form of pie charts. Using the ploty library of python. The data of users can also be downloaded in form of an excel(.csv format file) which provides the scope for further personalized analysis for specific purposes. The application can be best used by recruiters to get resumes from the applicants and they can sort the applicants very easily using our application. For example if the recruiter receiveshundredsofapplicationsthenmanuallychecking them could be a very cumbersome task through resume analyzertheprocesscouldbedoneinasmoothmanner.

4. EXPERIMENTAL SETUP

Web applications continue to evolve at an incredible rate, andthearchitecturearoundwebapplicationsisbecoming more and more complex. Most web applications rely on client-server architecture, where the client gives information and the server stores and retrieves the information. Most web applications available on the Internet are written in programming languages such as HTML, CSS for designing and animations, and Javascript used to create the front-end interface(client-side program). For scripting web applications, server-side programming is written using programming languages likePython,Java,PHP,andRuby,etc.PythonandJavahave commonly used languages for server-side programming. For storing the data from the web applications local databaseslikeMySQL,Oracle,PostgreSQL,etc.areused.A resumeanalyzerisa streamlit-basedwebapplicationthat analyzes an uploaded resume in the format of pdf and

extractsinformation.Thisanalysisisdonewiththehelpof pythonlibrariesthatincludepdfminer,nltk,spacy,pandas, NumPy,etc. The webapplicationhastwosections,mainly the normal user and admin panel. The user can upload his/her resume in the user section with a limit of pdf file upto 200 megabytes. Upon upload, the site user will be abletoseehis/herresumeandtheanalysisoftheresume which includes the user's current skills, recommended skills, resume score of the user , and the level of the user as per the analysis . This analysis helps the user to check forimprovementsinhis/herresumewhichcouldhelphim to get better opportunities. the resume data which is uploaded now gets stored in the local mysql database which can be accessed by the admin only through username and password. The admin section which unlocks through password has visual analysis of the total resume this analysis is represented in form of pie charts. Usingtheplotylibraryofpython.thedataofuserscanalso be downloaded inform ofanexcel(.csvformatfile)which provides the scope for further personalized analysis for specific purposes. The application can be best used by recruiterstogetresumesfromtheapplicantsandtheycan sort the applicants very easily using our application. For exampleiftherecruiterreceiveshundredsofapplications thenmanuallycheckingthemcouldbeaverycumbersome taskthroughresumeanalyzertheprocesscouldbedonein asmoothmanner.

5. RESULTS AND DISCUSSION

In this, we examined the Resume using NLP (Natural Language Processing) in which we interpreted information from a resume by identifying the main keywords and fabricating them into different sectors and finally suggesting the most applicable resumes to the HR or managers based on equivalent keywords. The interpreter parses the pivotal data from the resume and auto text a form for the user to interpret. After the confirmation, the resume is displayed to the users. Here, theusergetsacompleteoverviewoftheresumeintheGUI form. The interpreted data includes keywords such as education,certifications,workexperiences,socialprofiles.

International Research Journal of Engineering and Technology (IRJET)

e-ISSN:2395-0056

Volume: 09 Issue: 10 | Oct 2022 www.irjet.net p-ISSN:2395-0072

REFERENCES

[1] Ujjal Marjit, Kumar Sharma and Utpal Biswas, “ Discovering Resume Information Using Linked Data” , in International Journal of Web & Semantic Technology, Vol.3,No.2,April2012.

[2] Zhi Xiang Jiang, Chuang Zhang, Bo Xiao, Zhiqing Lin, “ Research and Implementation of Intelligent Chinese Resume Parsing” , WRI International Conference on CommunicationsandMobileComputing,Jan2009.

[3] ZhangChuang,WuMing,LiChunGuang,XiaoBo,“ Resume Parser: Semi-structured Chinese Document Analysis”,WRIWorldCongresson ComputerScienceand InformationEngineering,April2009.

Wefocusedontheemergingandfuture-changingjobsthat are available in the technical field. So the candidate must have the skills such as Data Science, Web Development, IOS Development, UI-UX Development, Android Development,etc.Ifthesesetofskillsarementionedinthe resumethenHe/Shewillgainabetterresumescore.Ifthe resume scoreis low then theapplication suggestsa set of skillsthatshouldbereflectedinHis/HerResume.

[4] CelikDuygu,KarakasAskyn,BalGulsen,Gultunca Cem, “Towards an Information Extraction System Based on Ontology to Match Resumes and Jobs” , IEEE 37th Annual Workshops on Computer Software and ApplicationsConferenceWorkshops,July2013.

[5] Di Wu, Lanlan Wu, Tieli Sun, Yingjie Jiang, “ Ontology based information extraction technology” , International Conference on Internet Technology and Applications(iTAP),Aug2011.

[6] Mayuri Verma, “Cluster basedRankingIndex for Enhancing Recruitment Process using Text Mining and Machine Learning” , International Journal of Computer Applications (0975 – 8887) Volume 157 – No 9, January 2017.

[7] Claudio Biancalana, Fabio Gasparetti, Alessandro Micarelli, Alfonso Miola, and Giuseppe Sansonetti, Contextaware movie recommendation based on signal processing and machine learning, Proceedings of the 2nd ChallengeonContext-AwareMovieRecommendation,p.510,October23-27,2011,Chicago,Illinois.

6. CONCLUSION

With a growing number of graduates every year, Getting placed in a job is one of the concerns of a student. A resume is a self-prepared synopsis of a person. It usually consists of one’s experiences and skills. A resume is usuallythefirstthing a recruiter oranorganization looks at to assess you. A strong resume leaves a strong first impression. Analyzing one’s resume gives us key insight into what it lacks therefore enabling us to modify and makeitbetter.Resumesareveryunderratedtoolstohelp youlandajob.Thisstreamlitbasedwebapplicationhopes to help graduates analyze their resumes and in turn help themgetplacedatajob.

[8] Feldman, Ronen, and James Sanger. The text mining handbook: advanced approaches in analyzing unstructureddata.CambridgeUniversityPress,2007.

[9] F.S. Gharehchopogh, |Approach and Review of User-Oriented Interactive Data Mining", IEEE, the 4th International Conference on Application of Information andCommunicationTechnologies(AICT2010),Uzbekistan, pp.1-4,2010.

[10] L. Huo, F. Yi, and H. Heping, 2008, "Dynamic Service Replica on Distributed Data Mining", 2008 IEEE International Conference on Computer Science and SoftwareEngineering,PP.390-393.

International Research Journal of Engineering and Technology (IRJET) e-ISSN:2395-0056

Volume: 09 Issue: 10 | Oct 2022 www.irjet.net p-ISSN:2395-0072

[11] Y. Shiqun, W. Gang, Q. Yuhui and Z. Weiqun, "Research and Implement of Classification Algorithm on Web Text Mining", 2007 IEEE, Third International Conference on Semantics Knowledge and Grid, 2007, pp. 446-449.

[12] Min-Ling Zhang; Zhi-Hua Zhou, Multilabel Neural Networks with Applications to Functional Genomics and TextCategorization,IEEETransactionsonKnowledgeand Data Engineering, Volume 18, Issue 10, Oct. (2006), Page(s):1338-1351

[13] Wai Lam, Miguel Ruiz, and Padmini Srinivasan. Automatic Text Categorization and Its Application to Text Retrieval, IEEE Transactions on Knowledge and Data Engineering. Volume 11,Issue 6 November (1999) Pages: 865879.

[14] TylerRichards-GettingStartedwithStreamlitfor Data Science_ Create and deploy Streamlit web applications from scratch in Python-Packt Publishing (2021).

[15] Akash, M. D. N., et al. "Check Certificate A Certificate Verification Platform for Students and Organizations." Smart Intelligent Computing and Applications, Volume 1. Springer, Singapore, 2022. 239245.