AI-DRIVEN CAREER GUIDANCE SYSTEM

Vaibhav Agarwal1
[Link] Gupta2
[Link] Agarwal3
1 U.G Scholar (vaibhavwork478@[Link]|9058015028) SRMSCET,Bareilly.
2 Assistant Professor ( ) SRMSCET,Bareilly.
3 Assistant Professor ( ) SRMSCET,Bareilly.

Abstract:
This research presents an AI-Enhanced Career Guidance System designed to provide
personalized and data-driven career recommendations for students, particularly those at the
10th-grade level. The system predicts suitable academic streams based on student abilities,
interests, and aptitude scores, and further recommends career pathways across 17 technical
domains using trained machine learning models. In addition, the system integrates automated
resume and cover-letter generation modules powered by natural-language models to support
students in professional documentation preparation. The objective of the proposed system is
to reduce decision-making uncertainty, bridge the information gap for young learners, and
empower students with personalized, scalable, and accessible career planning. Experimental
evaluations demonstrate strong model performance in classification accuracy, relevance
scoring, and user acceptance. The system holds significant potential for academic
institutions, e-learning platforms, and large-scale career counseling services.

Keyword:
Machine Learning Classification; Career Pathway Prediction; Stream Recommendation
System; Natural Language Processing (NLP); Resume and Cover-Letter Generation;
Educational Data Mining; Student Profiling; Aptitude and Interest Assessment; Multi-Class
Prediction Model; Random Forest Classifier; Transformer-Based NLP Models; Personalized
Career Counseling; Decision Support System; Data-Driven Career Prediction.

[Link]:
Career decision-making is a critical milestone for students, especially at the 10th-grade level,
where the choice of academic stream significantly influences future academic and
professional opportunities. Traditional career counseling in many regions remains limited,
subjective, or inaccessible. With advancements in artificial intelligence, data-driven
approaches can offer structured, reliable, and personalized guidance.
1

1
The proposed "AI-Enhanced Career Guidance System" utilizes machine learning and
natural-language processing to provide holistic career recommendations. The system
suggests suitable academic streams for 10th-grade students and predicts career pathways
across 17 technical fields—including Software Development, Cybersecurity, Data Science,
Cloud Computing, and others. Additionally, integrated resume and cover-letter generators
help students prepare professionally for internships, scholarships, or future opportunities.
1.1 Introduction Overview:

 Career selection after the 10th grade is a critical decision that significantly impacts a
student's future academic and professional opportunities.
 Traditional career counselling methods are often limited, subjective, and not
personalized to individual student needs.
 Advancements in Artificial Intelligence enable data-driven, scalable, and highly
personalized career guidance solutions.

[Link] Review:
AI-based career guidance systems have been widely studied across educational and
psychological domains, demonstrating strong potential in identifying student interests and
linking them to suitable career options. Existing research shows that machine learning
models can effectively utilize classification algorithms, psychometric indicators, cognitive
assessments, and aptitude-based evaluations to predict academic performance and
recommend skill pathways. Despite these advancements, many current systems still face
limitations, including narrow domain coverage, lack of personalization, reliance on outdated
rule-based mechanisms, and absence of automated professional document support.
Artificial intelligence has increasingly been adopted in the field of career counseling,
offering new possibilities for understanding student behaviour, interests, and aptitude
patterns. Prior research demonstrates that machine learning algorithms—such as decision
trees, neural networks, and probabilistic classifiers—can accurately analyze student data to
predict suitable academic and career directions. Many studies highlight the usefulness of
psychometric tests, cognitive ability assessments, and interest-based datasets in building
reliable career recommendation models.
Despite these advancements, existing solutions often lack comprehensive domain coverage
and fail to provide individualized guidance. Additionally, most prior models focus solely on
predicting career options and overlook essential components of career development, such as
preparing professional documents or offering multi-domain recommendations.
Recent literature emphasizes the need for integrative platforms capable of combining
aptitude evaluation, academic prediction, and personalized career mapping. To address these
gaps, the proposed system merges stream prediction, multi-domain technical career
classification, and NLP-based resume and cover-letter generation into a unified, intelligent
guidance framework. This combined approach enhances personalization, relevance, and
practical usability for students.
2
  Existing AI-based career guidance systems
 Recent research highlights the need for adaptive, data-driven solution.

[Link] Work:
The proposed work aims to develop an integrated AI-Enhanced Career Guidance System that
overcomes the limitations of traditional counselling by offering personalized, data-driven
academic and career recommendations. The system combines three major components—
stream prediction, technical career pathway classification, and automated professional
document generation—within a unified platform to support students during crucial
educational transitions.
The system collects student information such as academic scores, aptitude results, interest
indicators, and self-assessments, which are processed using machine learning models to
predict the most suitable academic stream for Grade 10 students. A multi-class classification
module further recommends career pathways across 17 technical domains based on student
skills and preferences.

Figure 1. Use Case Diagram of the AI-Enhanced Career Guidance System

The Use Case Diagram illustrates the interaction between users and the AI-Enhanced Career
Guidance System through two primary access levels: Unregistered Users and Registered
Users. Unregistered users can only view tool descriptions and must sign up to access the core
functionalities. Once registered, users can log in, manage their profiles, and utilize all AI-
powered tools.
3
After logging in, users gain access to the Core Tools Module, which includes:
 Career Prediction based on 17 skill scores,
 Automated Resume Generation in PDF format,
 Tailored Resume/Cover Letter Generation using job descriptions, and
 Stream Prediction based on academic marks.
The login use case acts as the central connector, enabling authenticated users to request
predictions or generate documents.
Overall, the diagram highlights how users interact with the system’s modules, ensuring a
clear flow from authentication to accessing intelligent career guidance features.

By utilizing machine learning-based stream classification, career pathway recommendation

models, and NLP-powered document generation, the proposed system delivers a unified,
personalized, and scalable solution. This integration ensures that students receive holistic
guidance and professionally aligned documents, ultimately improving clarity, reducing
decision fatigue, and supporting long-term academic and career success.

[Link] and Discussion:

The developed AI-Enhanced Career Guidance System successfully generated accurate and
personalized recommendations across all three major components: stream prediction, career
pathway selection, and NLP-based document generation. The machine learning model
demonstrated reliable classification performance, producing consistent stream predictions
that aligned with student aptitude and academic profiles. The multi-class career
recommendation module effectively mapped users to the most suitable technical domains
based on their interest and skill indicators. Additionally, the integrated NLP engine produced
clear, industry-ready resumes and cover letters with minimal user input. Overall, the system
4
delivered a smooth, user-friendly experience and proved to be an efficient and scalable
solution for supporting students in academic and career decision-making.
4.1 Experimental Setup:
 Data Collection: Gathered student academic records, aptitude scores, and skill/interest
profiles.
 Model Training: Developed machine learning classifiers for stream prediction and multi-
class career recommendation.
 System Evaluation: Tested end-to-end performance for accuracy, personalization, and
user experience.
4.2 Formula Setup:

5
 Figure [Link] Bar Chart Of Features.
[Link]:
This study presented a systematic evaluation of four distinct features (F1–F4) using a multi-
criteria efficiency model. Through quantitative analysis of efficiency percentages, response
times, and resource utilization, the features were rigorously differentiated. The
proposed Weighted Efficiency Score (WES) formula provided a balanced metric to rank the
features, revealing F3 as the optimal performer with 92.7% efficiency and the fastest
response time (88 ms). F2 followed closely, demonstrating strong efficiency (89.2%) with
better resource allocation. F1 and F4, while functional, exhibited notable limitations in speed
and consistency, respectively.
6
The use case diagram further contextualized these findings within a user-system interaction
framework, ensuring the evaluation aligns with practical workflows. The results validate the
applied methodology and offer a clear, data-driven pathway for feature selection and system
optimization. This work contributes a structured, repeatable evaluation framework suitable
for similar comparative analyses in software performance and resource management. Future
enhancements could integrate scalability assessments and real-user feedback to further refine
the decision-making model.

[Link]:
[1] M. Ahmad and A. Al-Shamri, "Career path prediction using deep neural networks," IEEE
Access, vol. 8, pp. 20110-20120, 2020.
[2] J. P. E. G. Orozco, L. A. P. Echeverri, and A. C. V. P. Giraldo, "Machine learning
techniques for career guidance: A comparative analysis," in Int. Conf. Soft Comput. Appl.,
2018, pp. 110–121.
[3] Z. Weng, G. Zhang, H. Yan, and Z. Liu, "Large language models in resume generation: A
comparative study of personalization and effectiveness," in Proc. 35th Annu. Conf. Innov.
Appl. AI, 2023, pp. 450-459.