Cosmetics Recommendation Using Decision Tree Classification Machine Learning Model
Cosmetics Recommendation Using Decision Tree Classification Machine Learning Model
Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR748
Cosmetics Recommendation Using Decision Tree
Classification Machine Learning Model
1
Narina Lokeswari; 2Sabbarapu Lavanya; 3Karumuri Harshitha; 4Parnandi Srinu Vasarao
Swarnandra College of Engineering and Technology
Abstract:- The interest for beauty care products has frequently. We can cross survey the outcome idea to our
developed as of late, particularly in the space of skincare, clients.
all over the planet. Customers have generally depended on
top selling things or ideas from the counter while shopping II. LITERATURE REVIEW
available. These is an inquiry that on the off chance that an
item will work with a specific client since everybody has a In ''Characterization In view of Choice Tree Calculation
particular skin condition. The primary objective of this for AI'' Behzad TahazJijol*, Adnan Mohsina Abdulazeez
proposition is to foster a framework for suggesting proposed the essential objective of the Choice tree calculation
skincare items in view of the client's skin type and the which is connected with the directed getting to know
cosmetics of the item. calculation family, is building a training model that can be
utilized to foresee the class or cost of objective factors by
Keywords:- Decision Trees, Personalized Beauty, Skin Type utilizing
Classification, Feature Extraction for Beauty Products,
Beauty Product Recommendation Systems, User Preferences Both Matsunami and Okuda utilized the technique for
in Cosmetics, Skin Undertone Analysis, Cosmetics Dataset, deciding client similitude to inspect audits of restorative Items.
Beauty Product Reviews. They removed positioning as well as genuine audits with
individual inclinations and Conclusions utilizing algorithmic
I. INTRODUCTION evaluating and k-implies group examination Ye likewise
utilized cooperative Separating, however she focused on
As a title state itself the procedure is used to recommend tending to the deficiencies of the ordinary methodology.
the restorative thing. As extra people began visiting the Evaluations are Utilized to affirm the outcomes despite the
excellence care items counter to get thing thoughts, there was fact that it doesn't channel things in view of them.
a looking at extension in the need for state of the art
developments. In any case, this technique is drawn-out and a To adjust skincare item suggestions very however much
significant part of the time incapable. Clients have found it as could be expected, Putriany utilized content-in view of their
attempting to go with the most ideal decisions as a result of the review. The framework was planned around the specific client.
staggering proportion of information that is open on the web. She focused on the client Record yet in addition considered
viewpoints like complexion, use, evaluating, portrayal, and
To determine the issue of information over-weight and photos for Better idea .To settle this issue, Jeong made a
smooth out the decision cycle, researchers have proposed a recommender framework that depends on the parts of beauty
couple recommender systems. Decision tree is used the most care products.
frequently. we can cross survey the outcome proposition to our
clients. III. METHODOLOGY
As a title state itself the system is used to recommend the This study aims to create a regression algorithm model
remedial thing. As extra people began visiting the excellence using the Google Collaboratory tool and Jupyter Notebook.
care items counter to get thing thoughts, there was a relating The steps of the modelling process are shown in Figure . The
extension in the essential for state of the art developments. first step is to obtain the datasets from the Kaggle website;
Regardless, this strategy is dreary and occasionally Then, the data is pre-processed, which involves Data Cleaning
unproductive. Clients have found it attempting to seek after and Feature Engineering; Next, the Exploratory Data Analysis
the ideal decisions in view of the staggering proportion of (EDA) is performed; To apply machine learning with the
information that is available on the web. Decision Tree and for some sample data is required. The table
below shows some data about different cosmetics products.
To determine the issue of information over-weight and The data is sourced from Kaggle.com .
smooth out the decision cycle, researchers have proposed a
couple recommender structures. Decision tree is used the most
IJISRT24MAR748 www.ijisrt.com 759
Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR748
Fig 1: Dataset
This data can cleaned and explorated using machine learning techniques and that mostly sutable for Decision tree and Random
forest algorithms .
Entropy
Entropy is the estimation of confusion or contaminations in the data handled in AI. It decides how a choice tree decides to divide
information.
IJISRT24MAR748 www.ijisrt.com 760
Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR748
Fig 3: General Architecture
Applying decision tree on the code:
IJISRT24MAR748 www.ijisrt.com 761
Volume 9, Issue 3, March – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24MAR748
Fig 4: Graphs for Different Skin Types
For this dataset we got the accuracy up to 84% by using [3]. Putriany, Villia, JaidanJauhari, and RahmatIzwanHeroza.
Decision making algorithm. “Item Clustering as An Input for Skin Care Product
Recommended System Using Content Based Filtering.”
IV. CONCLUSION Journal of Physics: Conference Series. Vol. 1196. No. 1.
IOP Publishing, 2019.
By leveraging algorithms such as decision tree, the [4]. Joanna Cristy Patty, Elika Thea Kirana, and Made
system aims to enhance the user experience by offering Sandra Diamond KhrismayantiGiri. 2018.
relevant and tailored product recommendations. Continuous Recommendations System for Purchase Of Cosmetics
monitoring, user feedback loops, and model updates are Using Content Based Filtering. International Journal of
crucial for maintaining the system's accuracy and relevance Computer Engineering and Information Technology 10,
over time. As the beauty industry evolves, incorporating 1 (Jan. 2018), 1–5.
emerging trends and staying attuned to user preferences are [5]. Jeong, Jiwon. “For Your Skin Beauty: Mapping
key considerations for ensuring the ongoing success of the Cosmetic Items with Bokeh.” (2018).
recommendation system. Ultimately, a well-designed and
maintained cosmetics recommendation system has the
potential to not only improve customer satisfaction but also
contribute to the growth and competitiveness of cosmetic
brands in the market.
REFERENCES
[1]. Hansson, Linda. “Product Recommendations in E-
commerce Systems using Content-based and
Collaborative Filtering.” (2015).Download (lu.se)
[2]. Ye, Hongwu. “A Personalized Collaborative Filtering
Recommendation Using Association Rules Mining and
Self-Organizing Map.” J. Softw. 6.4 (2011): 732-739.
