Scribd Logo
Upload

Welcome to Scribd!

  • 38 views

    Arooba Kanwal Assignment1

    Uploaded by

    Shehar Bano
    This document contains a final report submitted by Arooba Kanwal to Prof. Dr. Uzma Jamil for a Master of Philosophy in Computer Science. The report details 3 tasks completed for the assignment: [1] Exploring and downloading a freely available gender classification dataset from Kaggle, [2] Selecting 5 relevant research papers on the topic from the last 5 years, summarizing their techniques, results and limitations, [3] Stating the final problem statement, research questions and objectives. The dataset contains 5000 observations with 8 attributes and the papers explored techniques like machine learning classifiers, natural language processing and deep learning for gender classification with accuracies from 52-96.7% and noted opportunities for improved attributes and

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    Arooba Kanwal Assignment1

    Uploaded by

    Shehar Bano
    38 views4 pages
    This document contains a final report submitted by Arooba Kanwal to Prof. Dr. Uzma Jamil for a Master of Philosophy in Computer Science. The report details 3 tasks completed for the assignment: [1] Exploring and downloading a freely available gender classification dataset from Kaggle, [2] Selecting 5 relevant research papers on the topic from the last 5 years, summarizing their techniques, results and limitations, [3] Stating the final problem statement, research questions and objectives. The dataset contains 5000 observations with 8 attributes and the papers explored techniques like machine learning classifiers, natural language processing and deep learning for gender classification with accuracies from 52-96.7% and noted opportunities for improved attributes and

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document contains a final report submitted by Arooba Kanwal to Prof. Dr. Uzma Jamil for a Master of Philosophy in Computer Science. The report details 3 tasks completed for the assignment: [1] Exploring and downloading a freely available gender classification dataset from Kaggle, [2] Selecting 5 relevant research papers on the topic from the last 5 years, summarizing their techniques, results and limitations, [3] Stating the final problem statement, research questions and objectives. The dataset contains 5000 observations with 8 attributes and the papers explored techniques like machine learning classifiers, natural language processing and deep learning for gender classification with accuracies from 52-96.7% and noted opportunities for improved attributes and

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    38 views4 pages

    Arooba Kanwal Assignment1

    Uploaded by

    Shehar Bano
    This document contains a final report submitted by Arooba Kanwal to Prof. Dr. Uzma Jamil for a Master of Philosophy in Computer Science. The report details 3 tasks completed for the assignment: [1] Exploring and downloading a freely available gender classification dataset from Kaggle, [2] Selecting 5 relevant research papers on the topic from the last 5 years, summarizing their techniques, results and limitations, [3] Stating the final problem statement, research questions and objectives. The dataset contains 5000 observations with 8 attributes and the papers explored techniques like machine learning classifiers, natural language processing and deep learning for gender classification with accuracies from 52-96.7% and noted opportunities for improved attributes and

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 4

    Assignment No.

    1
    By

    Arooba Kanwal
    2220344

    Final report submitted

    To

    Prof. Dr. Uzma Jamil

    MASTER OF PHILOSOPHY
    IN
    COMPUTER SCIENCE

    DEPARTMENT OF COMPUTER SCIENCE


    GOVERNMERNT COLLAGE UNIVERSITY FAISALABAD

    Jun 2023
    Task 1: Explore and download a dataset which are freely available.
    Dataset Name: Gender Classification Dataset
    Size: The data contains 5000 observations with 8 attributes. (19KB)

    Type: gender_classification_v7.csv (text and numeric type)

    Source: Kaggle

    Task 2: Select research papers regarding dataset from last 5 years


    (2019-2023).
    Sr References Techniques/ methods Results Limitations/
    . Future work
    No
    1. (Vikas, Reddy, Data Pre-Processing: LogReg: 71.377 Future work for this
    C, & feature engineering, data XG Boost: 66.599 study can be done by
    Shanmugasund visualisation, and ML Multinomial NB: including additional
    aram, 2022) classifiers 68.570 attributes from the
    Gaussian NB:
    Training Machine dataset such as
    52.302
    Learning Models using RF: 68.303
    number of retweets,
    algorithms (Logistic language-based
    Regression, Random mining and many
    Forest, Gaussian NB, others which would
    Multinomial NB, enhance the accuracy
    XGBoost) and provide better
    results.
    2. (Vashisth & Natural Language LR (Logistic Future work for this
    Meehan, 2020) Processing (NLP) Regression) 54.43 study can be done by
    Compare multiple SVM (Support including additional
    techniques such as Bag of Vector Machine) attributes from the
    Words (TF-IDF), Word 52.67 dataset such as
    Embedding (W2Vec, Random Forest number of retweets,
    GloVe) and traditional 48.46 language-based
    Machine Learning XGBoost 52.39 mining and many
    techniques (Logistic others which would
    Regression, Support enhance the accuracy
    Vector Machine and and provide better
    Naïve Bayes) results.
    3. (Khan, Attique, CRFs based model is used The MSFS-CRFs We also plan to
    Syed, & Gul, to segment a face image. improved the improve the
    2019) Random Decision Forest average pixel performance of the
    (RDF) classifier, which labeling accuracy segmentation part by
    classifies the face images. (PLA) of the switching to the newly
    segmentation part introduced CNNs
    from 92.4% to based models.
    96.7%.
    4. (Wang, et al., Usage of various Natural Based on random Future work includes
    2022) Language Tool Kit forest classifier, expanding sarcasm
    (NLTK) libraries that we the got the detection to various
    process on textual accuracy 0.6683% modes (images,
    attributes of the dataset Based on Logistic GIFs), incorporating
    regression, we the social media data
    got the accuracy (emojis) for added
    0.5732% meaning, leveraging
    large datasets, and
    testing feature
    extraction methods,
    including word
    embedding in
    supervised and deep
    learning, to enhance
    sarcasm detection.
    5. (Adhinata & In this paper, human The method
    Junaidi, 2022) gender classification is achieves 90.49%
    addressed using the accuracy on JU-
    frontal and dorsal hand HD using
    images of a human Inception-v3

    Task 3: Final selection of dataset and research papers


    Problem Statement:

    Research Questions:

    Objectives:

    References
    Adhinata, F. D., & Junaidi, A. (2022). Gender Classifcation on Video Using FaceNet A lgorithm
    and Supervised Machine Learning. journal.uob.edu.bh.
    Khan, K., Attique, M., Syed, I., & Gul, A. (2019). Automatic Gender Classification through Face
    Segmentation. mdpi.
    Vashisth, P., & Meehan, K. (2020). Gender Classification using Twitter Text Data. IEEE Xplore.
    Vikas, K., Reddy, A. V., C, S. K., & Shanmugasundaram, H. (2022, July 04). User Gender
    Classification Based on Twitter Profile Using Machine Learning. IEEE Xplore.
    Wang, Z., Meng, Z., Saho, K., Uemura, K., Nojiri, N., & Meng, L. (2022). Deep learning-based
    elderly gender classification using Doppler radar. Springer.

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy