Upload 3 files

app.py

@@ -3,65 +3,9 @@ from langchain.llms import OpenAI
 import streamlit as st
 
 # Function to load OpenAI model and get responses
-def get_openai_response(question):
+def get_openai_response(question, resume_summary):
     # Concatenate resume details with the user's input
-    resume_details = """ DEEP MEHTA
-Rochester, NY | (720)-254-2220 | dm2399@rit.edu 
-GitHub - https://github.com/deepmehta922000
-LinkedIn - https://www.linkedin.com/in/deep-mehta-038ba91aa 
-
-SUMMARY
-Data Scientist proficient in Python, Java, R, and SQL, specializing in machine learning, deep learning, and algorithm optimization. Accomplished in developing and fine-tuning algorithms for data analysis. Eager to contribute expertise in driving data-driven decisions and solutions in a dynamic and challenging environment.
-
-EDUCATION
-Rochester Institute of Technology, Rochester | Master Data Science (GPA: 3.9/4.00) Expected May 2024
-Courses: Foundation Data Science, Advanced Statistics, Neural Networks, Relation Databases, Non-Relational Databases, Business Analytics and Intelligence
-
-SKILLS
-Programming Languages: Python, R, SQL, Java, C, MongoDB, XML
-Python Libraries: Pandas, NumPy, Keras, NLTK, SciPy, Matplotlib, Seaborn, TensorFlow, PyTorch
-Additional Tools: Tableau, PowerBI, RStudio, Git, AWS, Excel, Matlab, Minitab, JmpPro, OpenAI API
-Hard Skills: Data Analysis, Statistics, Database, Machine Learning, Deep Learning, Data Visualization, Quantitative Analysis, Big Data
-
-EXPERIENCE
-Lead Data Science Intern | S.A.K.E.C June 2021 – Aug 2021
-- Spearheaded the development of personalized course recommendations, reduced MAE by 12%, and increased precision by 15%.
-- Led a cross-functional team of developers in the implementation of collaborative, content-based, neural collaborative, and TF-IDF-enhanced NLP filtering algorithms reducing response time by 30%.
-- Employed Python libraries like Pandas and NumPy to preprocess and merge 7 datasets, reducing data discrepancies by 10%.
-
-Artificial Intelligence Intern | Digital Infrared Thermography Oct 2020 – June 2022
-- Developed diagnostic software using digital infrared thermography, image processing, and Neural Networks to detect breast.
-- Collaborated with researchers and medical professionals to implement algorithms like SVM, CNN, and Decision trees, resulting in a 92% accuracy.
-- Optimized image processing in MATLAB, reducing preprocessing time by 20% and enhancing overall image quality for comprehensive analysis.
-
-PROJECTS
-Diabetes Readmission Prediction with Machine Learning – Python June 2023
-- Engineered a diabetes patient readmission prediction model with an impressive precision score of 0.91 using sci-kit-learn, and XGBoost.
-- Trained classification models – Decision Trees, Randon Forest, KNN, and Logistic Regression using k-fold cross-validation to get the best model.
-- Utilized the GridSearch CV to find the best hyperparameters and tweak the model accordingly to give a 7% boost to the precision score.
-
-Data Analytics Consulting Virtual Intern | KPMG– Python July 2023
-- Assessed and successfully resolved more than a dozen data quality issues, leading to an improvement in data accuracy and completeness.
-- Demonstrated proficiency in utilizing data analytics tools, particularly Python and SQL enabling data-driven decision-making.
-- Designed an interactive dashboard using Tableau to showcase key performance metrics and data insights for stakeholders.
-
-British Airways: Customer Feedback Analysis & Sentiment Analysis – Python March 2023
-- Orchestrated end-to-end data scraping and cleaning pipelines, ensuring 100% data reliability and consistency.
-- Applied advanced NLP techniques like sentiment analysis, topic modeling, and text classification, extracting insights from customer feedback.
-- Leveraged tools and technology including Beautiful Soup, Scrapy, spaCy, seaborn, and sci-kit-learn, reducing project timelines by 20%.
-
-Liver Disease Prediction with Logistic Regression – Python Dec 2022
-- Employed Logistic Regression and evaluated the model on key performance metrics like precision (0.82), recall (0.94), and F1-measure (0.88).
-- Conducted thorough data preprocessing, applied advanced data imputation methods for handling missing values reducing data redundancy by a factor of 2.
-- Implemented under-sampling to address class imbalance, resulting in a more balanced dataset.
-
-PUBLICATIONS
-Applied Intelligence for Medical Diagnosing June 2022
-- Nikumbh, D. D., Sayyad, S., Mehta, D. V., Joshi, R. R., Dubey, K. S., & Matta, D. K. (2022). Applied Intelligence for Medical Diagnosing. In Handbook of Research on Applied Intelligence for Health and Clinical Informatics (pp. 44-79). IGI Global.
-
-
- """
-    input_prompt = f"{resume_details}\n\nUser Question: {question}"
+    input_prompt = f"{resume_summary}\n\nUser Question: {question}"
 
     llm = OpenAI(model_name="text-davinci-003", temperature=0.5)
     response = llm(input_prompt)
@@ -71,6 +15,10 @@ Applied Intelligence for Medical Diagnosing June 2022
 st.set_page_config(page_title="Q&A Demo")
 st.header("Chatbot")
 
+# Read resume information from the text file
+with open("resume.txt", "r", encoding="utf-8") as file:
+    resume_summary = file.read()
+
 # Initialize history list
 history = []
 
@@ -78,7 +26,7 @@ history = []
 input_question = st.text_input("Input: ", key="input")
 
 # Get response for the current question
-response = get_openai_response(input_question)
+response = get_openai_response(input_question, resume_summary)
 
 # Button to ask the question
 submit = st.button("Ask the question")

resume.txt

@@ -0,0 +1,55 @@
+DEEP MEHTA
+Rochester, NY | (720)-254-2220 | dm2399@rit.edu 
+GitHub - https://github.com/deepmehta922000
+LinkedIn - https://www.linkedin.com/in/deep-mehta-038ba91aa 
+
+SUMMARY
+Data Scientist proficient in Python, Java, R, and SQL, specializing in machine learning, deep learning, and algorithm optimization. Accomplished in developing and fine-tuning algorithms for data analysis. Eager to contribute expertise in driving data-driven decisions and solutions in a dynamic and challenging environment.
+
+EDUCATION
+Rochester Institute of Technology, Rochester | Master Data Science (GPA: 3.9/4.00) Expected May 2024
+Courses: Foundation Data Science, Advanced Statistics, Neural Networks, Relation Databases, Non-Relational Databases, Business Analytics and Intelligence
+
+SKILLS
+Programming Languages: Python, R, SQL, Java, C, MongoDB, XML
+Python Libraries: Pandas, NumPy, Keras, NLTK, SciPy, Matplotlib, Seaborn, TensorFlow, PyTorch
+Additional Tools: Tableau, PowerBI, RStudio, Git, AWS, Excel, Matlab, Minitab, JmpPro, OpenAI API
+Hard Skills: Data Analysis, Statistics, Database, Machine Learning, Deep Learning, Data Visualization, Quantitative Analysis, Big Data
+
+EXPERIENCE
+Lead Data Science Intern | S.A.K.E.C June 2021 – Aug 2021
+- Spearheaded the development of personalized course recommendations, reduced MAE by 12%, and increased precision by 15%.
+- Led a cross-functional team of developers in the implementation of collaborative, content-based, neural collaborative, and TF-IDF-enhanced NLP filtering algorithms reducing response time by 30%.
+- Employed Python libraries like Pandas and NumPy to preprocess and merge 7 datasets, reducing data discrepancies by 10%.
+
+Artificial Intelligence Intern | Digital Infrared Thermography Oct 2020 – June 2022
+- Developed diagnostic software using digital infrared thermography, image processing, and Neural Networks to detect breast.
+- Collaborated with researchers and medical professionals to implement algorithms like SVM, CNN, and Decision trees, resulting in a 92% accuracy.
+- Optimized image processing in MATLAB, reducing preprocessing time by 20% and enhancing overall image quality for comprehensive analysis.
+
+PROJECTS
+Diabetes Readmission Prediction with Machine Learning – Python June 2023
+- Engineered a diabetes patient readmission prediction model with an impressive precision score of 0.91 using sci-kit-learn, and XGBoost.
+- Trained classification models – Decision Trees, Randon Forest, KNN, and Logistic Regression using k-fold cross-validation to get the best model.
+- Utilized the GridSearch CV to find the best hyperparameters and tweak the model accordingly to give a 7% boost to the precision score.
+
+Data Analytics Consulting Virtual Intern | KPMG– Python July 2023
+- Assessed and successfully resolved more than a dozen data quality issues, leading to an improvement in data accuracy and completeness.
+- Demonstrated proficiency in utilizing data analytics tools, particularly Python and SQL enabling data-driven decision-making.
+- Designed an interactive dashboard using Tableau to showcase key performance metrics and data insights for stakeholders.
+
+British Airways: Customer Feedback Analysis & Sentiment Analysis – Python March 2023
+- Orchestrated end-to-end data scraping and cleaning pipelines, ensuring 100% data reliability and consistency.
+- Applied advanced NLP techniques like sentiment analysis, topic modeling, and text classification, extracting insights from customer feedback.
+- Leveraged tools and technology including Beautiful Soup, Scrapy, spaCy, seaborn, and sci-kit-learn, reducing project timelines by 20%.
+
+Liver Disease Prediction with Logistic Regression – Python Dec 2022
+- Employed Logistic Regression and evaluated the model on key performance metrics like precision (0.82), recall (0.94), and F1-measure (0.88).
+- Conducted thorough data preprocessing, applied advanced data imputation methods for handling missing values reducing data redundancy by a factor of 2.
+- Implemented under-sampling to address class imbalance, resulting in a more balanced dataset.
+
+PUBLICATIONS
+Applied Intelligence for Medical Diagnosing June 2022
+- Nikumbh, D. D., Sayyad, S., Mehta, D. V., Joshi, R. R., Dubey, K. S., & Matta, D. K. (2022). Applied Intelligence for Medical Diagnosing. In Handbook of Research on Applied Intelligence for Health and Clinical Informatics (pp. 44-79). IGI Global.
+
+Link - https://www.igi-global.com/chapter/applied-intelligence-for-medical-diagnosing/288808