app1.py

# -*- coding: utf-8 -*-
"""app1.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/12mKen8V8UQH6UcQ7d7jSUtOFeVt3ojqI
"""

import pandas as pd
import streamlit as st
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

# Function to calculate similarity scores between mentees and mentors
def calculate_similarity(mentee_df, mentor_df):
    # Combine mentee and mentor skills into lists of text (for vectorization)
    mentee_skills = mentee_df['Skills'].tolist()
    mentor_skills = mentor_df['Skills'].tolist()

    # Initialize TF-IDF vectorizer
    vectorizer = TfidfVectorizer(stop_words='english')

    # Combine all skills into a single list for fitting the vectorizer
    all_skills = mentee_skills + mentor_skills

    # Fit the vectorizer on all skills and transform them into vectors
    tfidf_matrix = vectorizer.fit_transform(all_skills)

    # Separate mentee and mentor skill vectors from the full matrix
    mentee_tfidf = tfidf_matrix[:len(mentee_skills)]
    mentor_tfidf = tfidf_matrix[len(mentee_skills):]

    # Calculate cosine similarity between each mentee and mentor pair
    similarity_scores = cosine_similarity(mentee_tfidf, mentor_tfidf)

    return similarity_scores

# Function to match mentees with mentors based on similarity scores
def match_mentor_mentee(mentees, mentors, similarity_scores):
    matches = []

    for i, mentee in mentees.iterrows():
        # Find the mentor with the highest similarity score for this mentee
        best_match_idx = similarity_scores[i].argmax()
        best_mentor = mentors.iloc[best_match_idx]

        matches.append({
            "Mentee Name": mentee['Name'],
            "Mentor Name": best_mentor['Name'],
            "Mentee Skills": mentee['Skills'],
            "Mentor Skills": best_mentor['Skills'],
            "Need": mentee['Need'],
            "Availability": best_mentor['Availability'],
            "Similarity Score": similarity_scores[i][best_match_idx]
        })

    return pd.DataFrame(matches)

# Streamlit UI
st.title("Mentor-Mentee Matching with Similarity Scores")

# Input for mentee data
st.header("Enter Mentee Data")
mentee_data = []
num_mentees = st.number_input("How many mentees?", min_value=1, max_value=10, value=1)
for i in range(num_mentees):
    mentee_name = st.text_input(f"Mentee Name {i+1}", key=f"mentee_name_{i}")
    mentee_skills = st.text_input(f"Mentee Skills {i+1}", key=f"mentee_skills_{i}")
    mentee_need = st.text_input(f"Mentee Need {i+1}", key=f"mentee_need_{i}")

    mentee_data.append([mentee_name, mentee_skills, mentee_need])

# Convert mentee data to DataFrame
mentee_df = pd.DataFrame(mentee_data, columns=["Name", "Skills", "Need"])

# Input for mentor data
st.header("Enter Mentor Data")
mentor_data = []
num_mentors = st.number_input("How many mentors?", min_value=1, max_value=10, value=1)
for i in range(num_mentors):
    mentor_name = st.text_input(f"Mentor Name {i+1}", key=f"mentor_name_{i}")
    mentor_skills = st.text_input(f"Mentor Skills {i+1}", key=f"mentor_skills_{i}")
    mentor_availability = st.selectbox(f"Mentor Availability {i+1}", ["Full-time", "Part-time"], key=f"mentor_availability_{i}")

    mentor_data.append([mentor_name, mentor_skills, mentor_availability])

# Convert mentor data to DataFrame
mentor_df = pd.DataFrame(mentor_data, columns=["Name", "Skills", "Availability"])

# If data is entered, proceed with matching
if st.button("Match Mentors and Mentees"):
    if not mentee_df.empty and not mentor_df.empty:
        # Calculate similarity scores
        similarity_scores = calculate_similarity(mentee_df, mentor_df)

        # Perform matching based on the highest similarity score
        matches_df = match_mentor_mentee(mentee_df, mentor_df, similarity_scores)

        if not matches_df.empty:
            matches_df.index = matches_df.index + 1
            st.write("Matching Results:")
            st.dataframe(matches_df)
            # Allow downloading of the result as a CSV file
            st.download_button(
                label="Download Matched Results",
                data=matches_df.to_csv(index=False),
                file_name="mentor_mentee_matches.csv",
                mime="text/csv"
            )
        else:
            st.write("No matches found based on the given criteria.")
    else:
        st.write("Please enter both mentee and mentor data.")