2.py

# -*- coding: utf-8 -*-
"""Job 23 Career matching algorithm.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1L1DhFFWrzWaQYEKFrAcvauarOGOw2EcU

CAREER MATCHING ALGORITHM

The goal is to develop a career matching algorithm.
Instead of just classification, we'll build a matching system based on;

*   Skill similarity
*   Skill importance
*   Required level
*   Threshold

Load and prepare data
"""

import pandas as pd

df = pd.read_excel("New jobready_career_skill_dataset_1000_rows_v2.xlsx")

# Remove duplicates
df = df.drop_duplicates()

df

"""Build career - Skills mapping"""

career_skills = df.groupby('Career').apply(
    lambda x: list(zip(
        x['Skill/Sub-skill'],
        x['Weight'],
        x['Threshold (0-100)'],
        x['Level Required']
    ))
).to_dict()

"""Define User Input"""

user_profile = {
    "skills": ["Python", "SQL", "Machine Learning", "Data Analysis"],
    "level": "Intermediate"
}

"""Matching Function"""

def match_career(user_profile, career_skills):
    scores = {}

    user_skills = set(user_profile["skills"])
    user_level = user_profile["level"]

    level_map = {
        "Beginner": 1,
        "Intermediate": 2,
        "Advanced": 3
    }

    for career, skills in career_skills.items():
        score = 0

        for skill, weight, threshold, level in skills:

            # Skill match
            if skill in user_skills:
                score += weight * 100   # boost score

            # Level match
            if level_map[user_level] >= level_map[level]:
                score += 5

            # Threshold influence
            score += (threshold / 100)

        scores[career] = score

    return sorted(scores.items(), key=lambda x: x[1], reverse=True)

"""Get Career Recommendations"""

results = match_career(user_profile, career_skills)

# Top 5 careers
top_careers = results[:5]

for career, score in top_careers:
    print(f"{career}: {round(score,2)}")

"""Interpretation

The algorithm recommends careers where the user’s skills best align with required skills, levels, and importance weights.

A career matching algorithm was developed to recommend suitable career paths based on user skills. The algorithm compares user input with career skill requirements using weighted scoring based on skill match, difficulty level, and threshold values. Careers are ranked according to similarity scores, and the system can also identify skill gaps and recommend learning resources. This approach provides a more practical and personalized solution compared to simple classification models.
"""