19IT446, 19IT448, 19IT471, 19IT474 Just Pics

A
PROJECT REPORT
ON
JustPics : Product Recommendation System

Submitted by
Yash Ghori (19IT446)
Meet Nakum(19IT448)
Rudra Trivedi(19IT471)
Divyarajsinh Rana(19IT474)

For Partial Fulfillment of the Requirements for Bachelor of Technology in Information

Technology

Guided by
Prof. Prachi Shah

December, 2022

Information Technology Department

Birla Vishvakarma Mahavidyalaya Engineering College
(An Autonomous Institution)
Vallabh Vidyanagar – 388120
Gujarat , India.

1
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Birla Vishvakarma Mahavidyalaya Engineering College

(An Autonomous Institution)
Information Technology Department
AY: 2022-23, Semester I

CERTIFICATE

This is to certify that the project work entitled JustPics: Product

Recommendation System has been successfully carried out by Yash
Ghori(19IT446), Meet Nakum(19IT448), Rudra Trivedi(19IT471),
Divyarajsinh Rana(19IT474) for the subject Project I (4IT31) during the
academic year 2022-23, Semester-I for partial fulfilment of Bachelor of
Technology in Information Technology. The work carried out during the semester
is satisfactory.

Prof. Prachi Shah Dr. Keyur Brahmbhatt

IT Department Head, IT Department
BVM
BVM

2
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

ACKNOWLEDGMENT

We have taken efforts in this project. However, it would not have been possible
without the kind support and help of many individuals. We would like to extend our
sincere thanks to all of them. We are highly thankful to our project coordinator Dr. Zankhna
Shah and our project guide Prof. Prachi Shah and for their guidance and constant supervision
as well as for providing necessary information regarding the project& also for their support in
completing the project. We also express thanks to Head of Department De. Keyur Brahmbhatt
for his involvement, constructive suggestion, and thoughtful idea for the betterment of the
project. We also thanks Dr. Indrajit Patel for providing us knowledge and all necessary
resources. We also thank our colleagues who have helped in the successful completion of the
project. At last, we would like to thank our parents who provide us moral support.

3
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

ABSTRACT

Most on-line shopping search engines are still largely depend on knowledge base and use
keyword matching as their search strategy to find the most likely product that consumers want
to buy.

Our web-app present a smart search engine for online shopping. Basically it uses images
as its input, and tries to understand the information about products from these image. First our
model use a neural network to extract features from the Image. Then use KNN( K Nearest
Neighbour) Algorithm to get the most similar images to the given input image. Euclidean
Distance is useful to calculate the similarity score for training data. Dataset Contains product
information data (including image, class label etc.) from different E-commerce websites like
Myntra and Ajio to learn these models. Specifically, our dataset contains information about
19000 products with image, and there are 20 categories in total. Finally we are able to
recommend products with similarity, and offer fast and accurate on-line shopping support.

4
 19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Index of Chapters

Chapter Name Page No.

Chapter 1: Introduction 8
Chapter 2: Analysis, Design Methodology, and Implementation Strategy 10
Chapter 3: Literature Review 24
Chapter 4: Implementation and Testing 28
Chapter 5: Conclusion and Future work 35
Chapter 6: References 36

5
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Index of Figures

Figure Name Page No.

Chapter 1

Figure 1.1 8

Chapter 2

Figure 2.1 10
Figure 2.2 12
Figure 2.3 13
Figure 2.4 13
Figure 2.5 15
Figure 2.6 16
Figure 2.7 17
Figure 2.8 18
Figure 2.9 19
Figure 2.10 20
Figure 2.11 21

Chapter 4

Figure 4.1 28
Figure 4.2 29
Figure 4.3 29
Figure 4.4 30
Figure 4.5 30
Figure 4.6 31
Figure 4.7 31
Figure 4.8 32

6
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Index of Tables

Table Page No.

Table 2.1 21
Table 4.1 33
Table 4.2 34
Table 4.3 34

7
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 1: Introduction

1.1. Aim:

This tool will work for recommending Similar Product by taking image as an input and
giving results with help of technologies like Image-Processing and Machine learning.

1.2. Project Scope:

The purpose of this tool is to get search result of the product faster than describing the
product in the search query. The system will use the different machine learning techniques like
image retrieval system, VggNet and models which are trained with thousands of images to
retrieve images which looks closest to the given input.

1.3. Project Objective:

Most on-line shopping search engines are still largely depending on knowledge base and
use key word matching as their search strategy to find the most likely product that consumers
want to buy. This is inefficient in a way that the description of products can vary a lot from the
seller’s side to the buyer’s side. we proposed an interactive product recommendation method,
which considers not only the product diversity but also the visual similarity, to interactively
capture a user's real intention and refine the product recommendation result based on the user's
real product interests.

Figure 1.1

8
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

1.4. Project Modules:

1.4.1. Login / Register

1.4.2. Search by Filters.
1.4.3. Capture / Upload photo
1.4.3.1. Applying Machine learning Model.
1.4.4. Display Results
1.4.5. User’s Past History
1.4.6. Redirect to the store.
1.4.7. Feedback and Ratings.

1.5. Project Requirements:

1.5.1. Hardware Requirement:

1.5.1.1. Desktop/Laptop with internet
1.5.2. Software Requirement:
1.5.2.1. Operating System: Windows 8 and Above
1.5.2.2. IDE: PyCharm
1.5.2.3. Programming Language: Python 3.9

9
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 2: Analysis, Design Methodology and

Implementation Strategy

2.1 Project Timeline chart:

Figure 2.1.

2.2 Detailed Module Description:

2.2.1 Login
Login Page for Users.

Register
New users can register themselves and make their account.

2.2.2 Search by Filters:

User can Search Images similar to their preferences. They can Filter
search results by different parameters or they can put more filters to get accurate
recommendations.

10
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.2.3 Capture / Upload photo:

Users can take the photo with device’s camera or they can upload
photo of the product directly from their device.

2.2.4 Display Results:

After applying the machine learning algorithms on the image the
results will be show to the user with different parameters which they can filter
the results with.

2.2.5 Redirecting to store:

If user will find their desired product, then they can go to the store
from Where the result was fetched and they can continue their shopping from
there.

2.2.6 Redirecting to store:

Customer who have registered their self on our website can give us their
precious feedback of any particular recommendation of product. It will help to
improve accuracy of our machine learning model.

11
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.3 Algorithm Design

Figure 2.2

Foremost all, in the beginning, we need to train our convolutional neural network with
the products dataset we have collected for the project. Data cleansing the other
activities is also required while making the dataset ready for the training. Once the
training is done the model can be used with the system to predict the product present
in the images.

12
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Figure 2.3

Web-portal is the main interface through which users can interact with the application.
As an input, the user can upload a new image and get recommendations as in which are other
product that is more likely related to the product present in the image. The “Recommendation
system” first finds the products which user may like based on similarity.

Figure 2.4

13
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.4 Language used

2.4.1 For Convolutional Neural Network development and training

• Python 3
• Libraries like
• Keras 1.2.2+
• TensorFlow R1.5
• Pandas
• NumPy
2.4.2 For Recommendation System
• Python 3

o For Connecting Backend to the Frontend

• Flask API

o For Web-Portal development

• HTML
• CSS
• React JS

14
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.5 SRS

2.5.1 Use Case Diagram

Figure 2.5.

In this diagram we can see how different users is going to use website. This shows the
relationship between users and the services. There are two types of users in this project
one is Admin and other is end-user. Admin will manage the website and properties and
end-user will buy/sell properties.

15
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.5.2 Data-Flow Diagram

11.2.1. Data-0

Figure 2.6.

16
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

11.2.2. Data-1

Figure 2.7.

DFD Level 1 provides a more detailed breakout of pieces of the Context Level
Diagram. You will highlight the main functions carried out by the system, as you break
down the high-level process of the Context Diagram into its subprocesses.

17
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

11.2.2. Data-2

Figure 2.8.

DFD Level 2 provides a more detailed breakout of pieces of the Context Level
Diagram. You will highlight the main functions carried out by the system, as you break
down the high-level process of the Context Diagram into its subprocesses.

18
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.5.3 Activity Diagram

Figure 2.9.

This is Activity Diagram which shows Different activity happened website at a time. This
shows the exact activity flow in website.

19
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.5.4 Class Diagram

Figure 2.10.

In this class diagram it displays the relationships between the objects, and to describe
what those objects do and the services that they provide.

20
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

2.5.5 E-R Diagram

Figure 2.11.

In this E-R diagram we can see the relationship between different entities. We can see
different Attributes of different entities. And how they in relation with other entity.

2.5.6 Data Dictionary

• User Table

Field Name Data Type Constraints Description

ID Int Primary Id
Password String Notnull Password
Username String Notnull Users name
Lastname String Notnull Users last name
Email String Notnull Users email
Firstname String Notnull Users first name
PastSearch Image Notnull Image of Past Search
PastResultID Int Notnull Result of past
Searches

Table

21
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Project Feasibility Study:

1. Operational feasibility:
- The whole purpose of this system is to handle the work much more accurately and
efficiently with less time consumption.
-The system is operationally feasible as it is very easy for the End users to operate it. -
-It only needs basic information about windows application.

2. Technical feasibility:

The technical needs of the system may include: Front-end and back-end selection An
important issue for the development of a project is the selection of suitable front-end and
back-end. When we decided to develop the project, we went through an extensive study to
determine the most suitable platform that suits the needs of the organization as well as helps
in development of the project. The aspects of our study included the following factors.

o Front-end selection: It must have a graphical user interface that assists users that are not
from IT background. So we have made front-end using React JS
GUI Features:
1. Scalability and extensibility.
2. Flexibility.
3. Easy to debug and maintain.

o Back-end Selection:
We have used Python as our Back-end Language which has the most widest library
collections
The technical feasibility is frequently the most difficult area encountered at this stage.
Our app
will fit perfectly for technical feasibility

22
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

3. Economic Feasibility:

The developing system must be justified by cost and benefit. Criteria to ensure that effort is
concentrated on project, which will give best, return at the earliest. One of the factors, which
affect
the development of a new system, is the cost it would require. Since the system is developed
as part of project work, there is no manual cost to spend for the proposed system. Also, all the
resources are already available, it gives an indication of the system is economically possible
for development.

23
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 3: Literature Review

Many Ecommerce websites like Amazon, Netflix use the personalized recommendation
system to promote their product and to push the user to but the products such as books, movies,
clothing and other consumer goods. Typical approaches followed in the recommendations are
Content Based Filtering and Collaborative Filtering. The Content Based recommendation
systems are used in the internet to have Web Personalizer. A mixture of both Content Based
Filtering and also Collaborative Filtering is used to recommend a new product. The previous
section discusses all the different approaches that is used to efficiently recommend the related
products to the users.

• The paper[1] uses the One Class Collaborative Filtering to recommend product in
the online forum. The image features are extracted using the Deep Convolution
Neural Network. Along with the extracted features from the image, user feedback
ratings and product evolution trends are also considered to recommend products to
the user.
• The paper[1] proposes a method to recommend handbags to each shopper. The
paper uses the Joint learning of attribute Projection and One Class SVM
classification based on the images of the shopper’s clicked images. The features of
the bag are extracted and are mapped to Projection Matrix. This projection matrix
is used in conjunction with SVM Classifier to have a new way of recommendation
system.
• The paper[1] proposes Image Based Recommendation System on Styles and
Substitutes. A method to identify the visually similar looking products are
recommended to the users in this paper.
• Visual Similarity is used to recommend the products in an online shopping forum.
Along with the product diversity, visual similarity is considered to capture the
intention of the user and refine the product recommendation system. [1]
• The paper[1] discusses about the recommendation system used in the Amazon.
Item-to-Item Collaborative Filtering is used to recommend the products in the
Amazon where the focus is to find the similar items and not the customers. For each
item purchased by the user and rated, the algorithm attempts to compute the similar
items associated with the item and would recommend the same to the user.

24
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

• The paper[1] focuses on improving the recommendation performance by learning

‘fashion aware’ by training the image representations and the recommendation
system jointly. The paper is similar to the work of Sisamese CNNs and was able to
show improvements compared to state of the art recommendation techniques such
as BPR and the variants that make use of pre-trained visual features.
• The paper[1] uses the convolution network to classify the input image to one of the
product categories. It uses another convolution network to identify the similar
product of the input image. Similarity is calculated using Jaccard Similarity.
• The paper[1] emphasizes the method to improve the performance of the product
recommendation with Sparse Data. Simplified Similarity Measure (SSM) is
proposed to handle the sparse data. It helps to reduce the problem of finding the
“dislike” group and thus helps in fast processing of the data.
• The paper[1] proposes the Deep Learning based Large Scale Visual
Recommendation. A unified Deep Convolution Neural Network architecture
VggNet is proposed in-order to learn embeddings to capture the notion of visual
similarity, across several semantic granularities.

25
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Related Work

We extend work on visually-aware recommendation with recent advances on representation

learning and generative models of images. We highlight a few of the main ideas from each
area below.

Recommender Systems:

At their core, recommender systems seek to model ‘compatibility’ between users and
items, based on historical observations of clicks, purchases, or interactions. Matrix
Factorization (MF) methods relate users and items by uncovering latent dimensions such that
users have similar representations to items they rate highly, and are the basis of many state-
of-the-art approaches. We are concerned with personalized ranking from implicit feedback
(e.g. distinguishing purchases from non-purchases, rather than estimating ratings), which can
be challenging due to the ambiguity of interpreting ‘non-observed’ (e.g. non-purchased) data.
Recently, point-wise and pairwise methods have successfully adapted MF to address such
challenges. Point-wise methods assume non-observed feedback to be inherently negative, and
cast the task in terms of regression, either by associating ‘confidence levels’ to feedback [13],
or by sampling non-observed feedback as negative instances [14]. Pairwise methods are
based on a weaker but possibly more realistic assumption that positive feedback must only be
‘more preferable’ than non-observed feedback. Such methods directly optimize the ranking
(in terms of the AUC) of the feedback and are to our knowledge state-of-the-art for implicit
feedback datasets. In particular, Bayesian Personalized Ranking (BPR), has experimentally
been shown to outperform a variety of competitive baselines [8]. Furthermore, BPRMF (i.e.,
BPR with MF as the underlying predictor) has been successfully extended to incorporate
(pre-trained) visual signals [1], and is thus the framework on which we build. Besides
images, others have developed content- and context aware models that make use of a variety
of information sources, including temporal dynamics [15], and geographic information [16],
[17]. Like our work these fall under the umbrella of ‘content aware’ recommender systems,
though are orthogonal due to the application domain and choice of signals considered.

26
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Deep Learning-Based Recommender Systems:

A variety of approaches have sought to incorporate ‘deep’ learning into recommender

systems [6], including systems that make use of content-based signals, such as for
recommendation of YouTube videos [18]. Our work is also a form of deep recommender
system, though is orthogonal to existing approaches in terms of data modality, except for a
few notable exceptions described below.

Visually-aware Recommender Systems:

Recent works have introduced visually-aware recommender systems where users’

rating dimensions are modelled in terms of visual signals in the system (product images).
Such visual dimensions were demonstrated to be successful at link prediction tasks, such as
recommending alternative (e.g. two similar t-shirts) and complementary (e.g. a t-shirt and a
matching pair of pants) items [12], and POI recommendation [19]. The most closely related
work is a recent method that extended BPR-MF to incorporate visual dimensions to facilitate
item recommendation tasks [1]. We build upon this framework, showing that substantially
improved performance can be obtained by using an ‘end-to-end’ learning approach (rather
than pre-trained features).

27
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 4: Implementation and Testing

User Interface

Home Page:

Fig. 4.1 Home Page

28
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Fig. 4.2 User Guide

Login Page:

Fig. 4.3 Login Page

29
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Register Page:

Fig. 4.4 Register Page

Forgot Password Page:

Fig. 4.5 Forgot Password Page

30
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Upload Photo:

Fig. 4.6 Upload Photo

Displaying the Result:

Fig. 4.7 Displaying Result

31
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

History of the User:

Fig. 4.8 History Of User

32
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Data Set:
Sample for Ajio:

Product_URL
Brand DescriptionId_ProductURL_imageCategory_by_gender Discount Price
Original
(in Rs.)
Price
Color
(in Rs.)
https://www.ajio.com/netplay-checked-polo-t-shirt/p/441137362_white
netplay Checked Polo 4.41E+11
T-shirt https://assets.ajio.com/medias/sys_master/root/20220309/inTn/6227b81f
Men 559 699 white
https://www.ajio.com/netplay-tapered-fit-flat-front-trousers/p/441124497_navy
netplay Tapered Fit 4.41E+11
Flat-Front https://assets.ajio.com/medias/sys_master/root/20210907/vQKt/6136775c
Trousers Men 720 1,499 navy
https://www.ajio.com/the-indian-garage-co-striped-slim-fit-shirt-with-patch-pocket/p/460453612_white
the-indian-garage-co
Striped Slim Fit
4.6E+11
Shirt with
https://assets.ajio.com/medias/sys_master/root/20211228/s1km/61ca36a
Patch Pocket
Men 495 1,649 white
https://www.ajio.com/performax-heathered-crew-neck-t-shirt/p/441036730_charcoal
performax Heathered Crew-Neck
4.41E+11 https://assets.ajio.com/medias/sys_master/root/20220120/lLur/61e981ae
T-shirt Men 329 599 charcoal
https://www.ajio.com/john-players-jeans-washed-skinny-fit-jeans-with-whiskers/p/441128531_jetblack
john-players-jeans
Washed Skinny 4.41E+11
Fit Jeans
https://assets.ajio.com/medias/sys_master/root/20210728/I1Im/61005afff
with Whiskers
Men 899 999 jetblack
https://www.ajio.com/dennislingo-premium-attire-slim-fit-shirt-with-patch-pocket/p/462323964_olive
dennislingo-premium-attire
Slim Fit Shirt4.62E+11
with Patch https://assets.ajio.com/medias/sys_master/root/20210421/ZxwY/607fbcdb
Pocket Men 684 1,849 olive
https://www.ajio.com/dennislingo-premium-attire-striped-slim-fit-shirt/p/461119026_black
dennislingo-premium-attire
Striped Slim4.61E+11
Fit Shirt https://assets.ajio.com/medias/sys_master/root/20210403/ywxs/6068696
Men 700 2,499 black
https://www.ajio.com/teamspirit-track-pants-with-contrast-taping/p/441031716_ltblue
teamspirit Track Pants 4.41E+11
with Contrast
https://assets.ajio.com/medias/sys_master/root/hfd/h39/15879850000414
Taping Men 419 499 ltblue
https://www.ajio.com/performax-fastdry-active-crew-neck-t-shirt/p/441128501_jetblack
performax Fastdry Active 4.41E+11
Crew-Neckhttps://assets.ajio.com/medias/sys_master/root/20220125/5ENA/61eef65
T-shirt Men 374 499 jetblack
https://www.ajio.com/ardeur-pack-of-2-shorts-with-insert-pocket/p/461730457_black
ardeur Pack of 2 Shorts
4.62E+11
with Insert
https://assets.ajio.com/medias/sys_master/root/20220519/oL6x/62863a81
PocketMen 500 1,998 black
https://www.ajio.com/the-indian-garage-co-checked-slim-fit-shirt-with-patch-pocket/p/460698898_navy
the-indian-garage-co
Checked Slim 4.61E+11
Fit Shirt with
https://assets.ajio.com/medias/sys_master/root/20210316/TIPl/604faba1f
Patch Pocket
Men 490 1,749 navy
https://www.ajio.com/campus-sutra-striped-regular-fit-shirt/p/463611345_white
campus-sutra
Striped Regular
4.64E+11
Fit Shirthttps://assets.ajio.com/medias/sys_master/root/20220425/tUDF/626631e
Men 675 1,499 white
https://www.ajio.com/teamspirit-heathered-crew-neck-t-shirt/p/441000453_red
teamspirit Heathered Crew-Neck
4.41E+11 https://assets.ajio.com/medias/sys_master/root/h97/h0d/1348045491407
T-shirt Men 335 399 red
https://www.ajio.com/netplay-leaf-print-shirt-with-patch-pocket/p/441136451_mintgreen
netplay Leaf Print Shirt 4.41E+11
with Patch
https://assets.ajio.com/medias/sys_master/root/20220224/uLAB/62168f18
PocketMen 849 999 mintgreen
https://www.ajio.com/netplay-leaf-print-polo-t-shirt/p/441137352_olive
netplay Leaf Print Polo 4.41E+11
T-shirt https://assets.ajio.com/medias/sys_master/root/20220301/DDoY/621e35e
Men 559 699 olive
https://www.ajio.com/netplay-slim-fit-flat-front-chinos-with-insert-pockets/p/441121060_green
netplay Slim Fit Flat-Front
4.41E+11
Chinos
https://assets.ajio.com/medias/sys_master/root/20210731/nfti/61044ac9a
with Insert
MenPockets 779 1,299 green
https://www.ajio.com/netplay-mid-rise-flat-front-cropped-slim-fit-chinos/p/441121060_ltgrey
netplay Mid-Rise Flat-Front
4.41E+11Cropped
https://assets.ajio.com/medias/sys_master/root/20210803/yCZG/610844f1
SlimMen
Fit Chinos 779 1,299 ltgrey
https://www.ajio.com/teamspirit-heathered-polo-t-shirt/p/441000674_maroon
teamspirit Heathered Polo4.41E+11
T-shirthttps://assets.ajio.com/medias/sys_master/root/hf3/h40/15590223183902
Men 503 599 maroon
https://www.ajio.com/netplay-mid-rise-shorts-with-insert-pockets/p/440929331_navy
netplay Mid-Rise Shorts 4.41E+11
with Insert
https://assets.ajio.com/medias/sys_master/root/h12/h80/1225916085046
Pockets
Men 212 299 navy

33
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Sample for Flipkart:

uniq_id crawl_timestamp
product_urlproduct_name
product_category_tree retail_pricediscounted_price
image description brand
c2d766ca982eca8304150849735ffef9
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2ffvzetthbb?pid=SRTEH2FF9KEDEFGF
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
ShortsClothing
999 >> Lingerie,
379 ["http://img5a.flixcart.com/image/short/u/4/a/altht-3p-21-
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
7f7036a6d550aaa89d34c77bd39a5e48
2016-03-25http://www.flipkart.com/fabhomedecor-fabric-double-sofa-bed/p/itmeh3qgfamccfpy?pid=SBEEH3QGU7MFYJFY
22:59:23 +0000
FabHomeDecor["Furniture
Fabric >>
Double
LivingSofa
Room
32157
BedFurniture
22646
>> Sofa
["http://img6a.flixcart.com/image/sofa-bed/j/f/y/fhd112-do
Beds &FabHomeDecor
Futons >> FabHomeDecor
Fabric
FabHomeDecor
DoubleFabric
SofaDouble
Bed (Finish
Sofa Co
Be
f449ec65dcbc041b6ae5e6a32717d01b
2016-03-25http://www.flipkart.com/aw-bellies/p/itmeh4grgfbkexnt?pid=SHOEH4GRSUBJGZXE
22:59:23 +0000
AW Bellies ["Footwear >> Women's 999 Footwear >> 499Ballerinas
["http://img5a.flixcart.com/image/shoe/7/z/z/red-as-454-a
>> Key
AW Features
Bellies"] of AWAW Bellies Sandals Wedges Heel C
0973b37acd0c664e3de26e97e5571454
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2f6sdgah2pq?pid=SRTEH2F6HUZMQ6SJ
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
ShortsClothing
699 >> Lingerie,
267 ["http://img5a.flixcart.com/image/short/6/2/h/altght-11-ali
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
bc940ea42ee6bef5ac7cea3fb5cfbee7
2016-03-25http://www.flipkart.com/sicons-all-purpose-arnica-dog-shampoo/p/itmeh3zyw2vhgsp5?pid=PSOEH3ZYDMSYARJ5
22:59:23 +0000
Sicons All Purpose
["Pet Supplies
Arnica >>
DogGrooming
Shampoo220>> Skin &210
Coat
["http://img5a.flixcart.com/image/pet-shampoo/r/j/5/sh-df
Care >> Shampoo
Specifications
>> Sicons
of Sicons
Sicons
All Purpose
All Purpose
ArnicaArnica
Dog Dog
Shamp Sh
c2a17313954882c1dba461863e98adf2
2016-03-25http://www.flipkart.com/eternal-gandhi-super-series-crystal-paper-weights-silver-finish/p/itmeb7h2hrfgutxb?pid=PWTEB7
22:59:23 +0000
Eternal Gandhi
["Eternal
SuperGandhi
Series Crystal
Super Series
Paper
430 Crystal
Weights
430
Paper
with
["http://img5a.flixcart.com/image/paper-weight/u/e/3/eter
Weight..."]
Silver Finish
Key Features of Eternal
Eternal Gandhi
GandhiSuper Series Cryst
ce5a6818f7707e2cb61fdcdbba61f5ad
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2ftwkzykhcg?pid=SRTEH2FVVKRBAXHB
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
Shorts1199
Clothing >> Lingerie,
479 ["http://img6a.flixcart.com/image/short/p/j/z/altght4p-26-a
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
8542703ca9e6ebdf6d742638dfb1f2ca
2016-03-25http://www.flipkart.com/fabhomedecor-fabric-double-sofa-bed/p/itmeh3qgnxztzcvf?pid=SBEEH3QGYGHFUEXN
22:59:23 +0000
FabHomeDecor["Furniture
Fabric >>
Double
LivingSofa
Room
32157
BedFurniture
22646
>> Sofa
["http://img6a.flixcart.com/image/sofa-bed/e/x/n/fhd107-d
Beds &FabHomeDecor
Futons >> FabHomeDecor
Fabric
FabHomeDecor
DoubleFabric
SofaDouble
Bed (Finish
Sofa Co
Be
29c8d290caa451f97b1c32df64477a2c
2016-03-25http://www.flipkart.com/dilli-bazaaar-bellies-corporate-casuals-casuals/p/itmeh2paagfuhbzh?pid=SHOEH3DZBFR88SCK
22:59:23 +0000
dilli bazaaar["Footwear
Bellies, Corporate
>> Women'sCasuals,
699
Footwear
Casuals>>
349Ballerinas
["http://img6a.flixcart.com/image/shoe/b/p/n/pink-200db2
>> Key
dilli Features
bazaaar Bellies,
of dillidilli
bazaaar
Corporate
bazaaar Bellies,
Casuals,
Corporate
Casuals"]
Ca
4044c0ac52c1ee4b28777417651faf42
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2fvdphhyshh?pid=SRTEH2FVUHAAVH9X
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
Shorts1199
Clothing >> Lingerie,
479 ["http://img5a.flixcart.com/image/short/5/z/c/altght4p-39-
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
e54bc0a7c3429da2ebef0b30331fe3d2
2016-03-25http://www.flipkart.com/ladela-bellies/p/itmeh4kmxght7tuc?pid=SHOEH4KM2W3Z6EH5
22:59:23 +0000
Ladela Bellies
["Footwear >> Women's1724 Footwear >> 950Ballerinas
["http://img5a.flixcart.com/image/shoe/s/g/m/black-r998-2
>> Key
LadelaFeatures
Bellies"]
of Ladela
Ladela Bellies Brand: LADELA Colo
c73e78fb440ff8972e0762daed4fc109
2016-03-25http://www.flipkart.com/carrel-printed-women-s/p/itmehf3ednm5fvyt?pid=SWIEHF3EF5PZAZUY
22:59:23 +0000
Carrel Printed
["Clothing
Women's >> Women's 2299 Clothing >> Sports
910 ["http://img6a.flixcart.com/image/swimsuit/5/v/9/carrel-sw
& Gym Wear Key Features
>> Swimsuits
of Carrel
>>
Carrel
Carrel
Printed
Swimsuits
Women's >> Fabric:
Carrel PrSw
c29af37837afcaf44b779eca7c19295f
2016-03-25http://www.flipkart.com/sicons-all-purpose-tea-tree-dog-shampoo/p/itmeh3zyhfrgh3re?pid=PSOEH3ZYYFETGCCD
22:59:23 +0000
Sicons All Purpose
["Pet Supplies
Tea Tree
>> Dog
Grooming
Shampoo>> Skin & Coat["http://img6a.flixcart.com/image/pet-shampoo/c/c/d/sh-d
Care >> Shampoo
Specifications
>> Sicons
of Sicons
Sicons
All Purpose
All Purpose
Tea Tree
Tea Tree
Dog Dog
Sham
9aacdecceb404c74abddc513fd2756a8
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2fgcjz4mzvu?pid=SRTEH2FGBDJGX8FW
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
ShortsClothing
999 >> Lingerie,
379 ["http://img6a.flixcart.com/image/short/q/z/v/altht-3p-17-a
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
83c53f8948f508f51d2249b489ca8e7d
2016-03-25http://www.flipkart.com/freelance-vacuum-bottles-350-ml-bottle/p/itmegytzpkgmrew5?pid=BOTEGYTZ2T6WUJMM
22:59:23 +0000
Freelance Vacuum
["Pens &Bottles
Stationery
350 ml>> Bottle
School
699 Supplies699>>["http://img5a.flixcart.com/image/bottle/j/m/m/av004bgr-f
Water Bottles
Specifications
>> Freelance
of Freelance
Freelance
Water Bottles Vacuum>> Freelance
Bottles 350Vacm
d95b0456a0350bc42f2393c6e84b0f09
2016-03-25http://www.flipkart.com/alisha-solid-women-s-cycling-shorts/p/itmeh2fehghynve9?pid=SRTEH2FECMGNZJXJ
22:59:23 +0000
Alisha Solid["Clothing
Women's >> Cycling
Women's
ShortsClothing
999 >> Lingerie,
379 ["http://img5a.flixcart.com/image/short/t/p/b/altht-3p-2-al
Sleep &KeySwimwear
Features>>ofShorts
Alisha
Alisha >>
Solid
Alisha
Women's
Shorts Cycling
>> Alisha
Sho So
849ab05698081a374215d0b7d18047d0
2016-03-25http://www.flipkart.com/fabhomedecor-fabric-double-sofa-bed/p/itmeh3qgzgmsuyxn?pid=SBEEH3QGAYAEPRCG
22:59:23 +0000
FabHomeDecor["Furniture
Fabric >>
Double
LivingSofa
Room
32157
BedFurniture
22646
>> Sofa
["http://img5a.flixcart.com/image/sofa-bed/r/c/g/fhd132-d
Beds &FabHomeDecor
Futons >> FabHomeDecor
Fabric
FabHomeDecor
DoubleFabric
SofaDouble
Bed (Finish
Sofa Co
Be
c275ee5ac19f774a3ef7da71b40aabd8
2016-03-25http://www.flipkart.com/style-foot-bellies/p/itmeh4fssgzabe5h?pid=SHOEH4FSHEDEMVTM
22:59:23 +0000
Style Foot Bellies
["Footwear >> Women's 899 Footwear >> 449Ballerinas
["http://img5a.flixcart.com/image/shoe/p/f/h/black-sfwf03
>> Key
StyleFeatures
Foot Bellies"]
of Style
Style FootFootBellies ballerina shoes
4f3511c33a6869b1d5102cd97818ef00
2016-03-25http://www.flipkart.com/carrel-printed-women-s/p/itmehf3hmxnuawtn?pid=SWIEHF3H7WG6VGMR
22:59:23 +0000
Carrel Printed
["Clothing
Women's >> Women's 2499 Clothing >> Sports
999 ["http://img5a.flixcart.com/image/swimsuit/g/z/j/carrel-sw-
& Gym Wear Key Features
>> Swimsuits
of Carrel
>>
Carrel
Carrel
Printed
Swimsuits
Women's >> Fabric:
Carrel PrSw

Sample for Myntra:

uniq_id crawl_timestamp
product_idlink size brand care_instructionsactual_colorproduct_type images product_details type variant_pricevariant_compare_at_price
ideal_for
cc51c3dd925d9d84e2fb28c7918dbbc3
2019-08-11 05:49:46
6937673 +0000
https://www.myntra.com/tops/imara/imara-women-black-solid-top/6937673/buy
XL IMARA Polyester |Black Dry-clean Top http://assets.myntassets.com/v1/assets/images/6937673/2018/8/30/88c6ad02-eab9-42c8-
Black solid Clothing/Women/Tops/IMARA/More
woven regular top,has
959 a V-neck,
1599 three-quarter
Women by IMARAsleeves, an attached lin
c7e2fe388a93dae5c86b50dd8a5dbd7a
2019-07-03 01:04:34
7441182 +0000
https://www.myntra.com/kurtas/house-of-pataudi/house-of-pataudi-men-black-printed-straight-kurta/7441182/buy
XL House of Pataudi
100% cottonMachine-wash
Black Straight Kurta http://assets.myntassets.com/v1/assets/images/7441182/2018/11/10/7f473119-8e08-4548
Black printed Clothing/Men/Kurtas/House
straight kurta, has
799a mandarin
1999
of Pataudi/More
collar,
Men long sleeves, by Housestraight
of Pataudi
hem, side
76949d78057a633aace6246045eefd66
2019-07-15 09:43:31
9245141 +0000
https://www.myntra.com/pillows/the-white-willow/the-white-willow-unisex-off-white-therapedic-memory-foam-pillow/9245141/buy
S The White Willow
PolycottonWhite
| Only spot clean by hand http://assets.myntassets.com/v1/assets/images/9245141/2019/4/10/c4aecdbc-1c39-4c4b-
with warm
Set content:
water Home/Unisex/Pillows/The
and
1 pillow
mild detergent
Colour:
1399
Off
| For
white
removing
White
2799
Filling:
Unisex
Willow/More
stains
Memory or odor
foambysprinkle
The
Type:
White
Therapedic
baking
Willow
powdM
5f83f6b51da1ef93bc0585577b6991e6
2019-07-15 14:11:14
1314889 +0000
https://www.myntra.com/dupatta/dupatta-bazaar/dupatta-bazaar-pink-dupatta/1314889/buy
Onesize Dupatta Bazaar
Chiffon | Hand-wash
Pink Dupatta http://assets.myntassets.com/v1/assets/images/1314889/2016/4/20/11461153070320-Dup
Pink dupatta with
Clothing/Women/Dupatta/Dupatta
crinkled
349 effect 499 Women Bazaar/More by Dupatta Bazaar
890bf4355c319a4211296b9d3176d11f
2019-06-01 08:17:28
7705322 +0000
https://www.myntra.com/kurta-sets/manyavar/manyavar-men-yellow--white-self-design-kurta-with-churidar/7705322/buy
S Manyavar Top fabric: Yellow
Silk Blend
| White
|Kurta
Bottomwithfabric:
Churidar
http://assets.myntassets.com/v1/assets/images/7705322/2018/10/29/e9ea701d-93aa-460f
Silk Blend
Yellow| Dry-clean
andClothing/Men/Kurta
white self design2999
kurta
Sets/Manyavar/More
with2999
churidarYellow
Men bystraight Manyavarknee length kurta,
6cb35dc135630e138f618b455014bf27
2019-07-28 13:03:52
2160958 +0000
https://www.myntra.com/kurta-sets/wintage/wintage-men-beige-solid-kurta-with-pyjamas/2160958/buy
L Wintage Kurta fabric:Beigelinen | Bottom
Kurta with
fabric:Pyjamas
http://assets.myntassets.com/v1/assets/images/2160958/2017/10/5/11507205292287-Win
pure cotton
Beige| Dry-clean
solid Clothing/Men/Kurta
kurta with pyjamas979Beige
Sets/Wintage/More
straight
1399 Men knee length
by Wintage
kurta, has a mandarin co
07cc95890eb0034fda8c7160a3a5be6d
2019-06-17 05:21:59
7766518 +0000
https://www.myntra.com/skirts/geroo-jaipur/geroo-women-green-cotton-hand-block-print-skirt/7766518/buy
34 Geroo JaipurFirst wash dry
Greenclean than after hand http://assets.myntassets.com/v1/assets/images/7766518/2018/11/12/ec7ea490-1f0a-47a2
wash Green coloured Clothing/Women/Skirts/Geroo
free-size dabu
1572print cotton
1850Jaipur/More
skirt
Women for womenby Geroo
with drawstring
Jaipur closu
0b46713a5d8c659199965af7ad64ddca
2019-07-15 04:00:15
1923426 +0000
https://www.myntra.com/kurtas/the-indian-garage-co/the-indian-garage-co-men-blue--white-printed-a-line-kurta/1923426/buy
XL The Indian Garage
Cotton Co| Machine-wash
Blue | WhiteA-Line Kurtahttp://assets.myntassets.com/v1/assets/images/1923426/2017/6/21/11498040802877-The
Blue and whiteClothing/Men/Kurtas/The
printed A-line479kurta, has
Indian
1199
a mandarin
Garage
Men collar,
Co/More longbysleeves,
The Indian
zip closure
Garag
8ec62e9d32f52999eda6e2bccb777218
2019-06-17 00:52:43
6551403 +0000
https://www.myntra.com/kurtas/anouk/anouk-women-maroon-solid-a-line-kurta/6551403/buy
XS Anouk Polyester |Maroon Dry-clean A-Line Kurtahttp://assets.myntassets.com/v1/assets/images/6551403/2018/12/19/1ab6bfba-4825-4e59
Maroon solid Clothing/Women/Kurtas/Anouk/More
A-line kurta, has
1359a round3399
neck,Women
three-quarter
by Anouksleeves, flared hem
8cc3d10b36c41a23f579ccb36c58d211
2019-07-30 07:24:13
8389801 +0000
https://www.myntra.com/kurta-sets/salwar-studio/salwar-studio-boys-orange--white-printed-kurta-with-pyjamas/8389801/buy
3-4Y SALWAR STUDIO
Top fabric: White
Pure Cotton
| Orange
Kurta
| Bottom
with Pyjamas
http://assets.myntassets.com/v1/assets/images/productimage/2019/1/4/524faa41-cecd-48
fabric: PureOrange
Cottonand| Clothing/Boys/Kurta
Hand-wash
white printed kurta
586with
Sets/SALWAR
pyjamas
1955 Boys Orange
STUDIO/More
straightbyabove
SALWARkneeSTUDIO
kurta, h
353fff7925a669be503c221c317ec987
2019-08-06 13:58:15
9791991 +0000
https://www.myntra.com/palazzos/clora-creation/clora-creation-women-navy-blue-wide-leg-printed-palazzos/9791991/buy
40 Clora Creation
Viscose Rayon
Blue||Hand-wash
NavyPrinted Palazzos
http://assets.myntassets.com/v1/assets/images/productimage/2019/5/24/cd9e6d5d-50cf-4
A pair of navyClothing/Women/Palazzos/Clora
blue printed woven
979 wide1399 leg palazzos
Women
Creation/More by Clora Creation
10fd08b921bc9ec918e6305eb979e000
2019-07-25 01:41:08
7098789 +0000
https://www.myntra.com/kurtas/libas/libas-women-blue--white-striped-straight-kurta/7098789/buy
S Libas Rayon | Hand-wash
Blue | WhiteStraight Kurta http://assets.myntassets.com/v1/assets/images/7098789/2018/8/20/dad722b8-f2c0-4e7e-
Blue and whiteClothing/Women/Kurtas/Libas/More
striped straight
519kurta, has
1299a shirt
Women collar
by Libas
with half hook-and-eye pla
ecf22e7affda59a926112300a3bebf9f
2019-08-11 00:45:18
7098851 +0000
https://www.myntra.com/kurta-sets/libas/libas-women-coffee-brown-solid-kurta-with-palazzos/7098851/buy
M Libas Top fabric: Brown
Tussar |Silk
Coffee
Kurta
| Bottom
Brown
withfabric:
Palazzos
http://assets.myntassets.com/v1/assets/images/7098851/2018/8/17/8728af1c-969f-445c-b
TussarCoffee
Silk | brown
Hand-wash
Clothing/Women/Kurta
solid kurta with
1119palazzos
Sets/Libas/More
2799
Coffee Women
brownbystraight
Libas calf length kurta, ha
b9ee69961a8ad431a09993bb0353b4e2
2019-07-02 11:48:00
1681450 +0000
https://www.myntra.com/kurtas/anouk/anouk-women-navy-printed-straight-kurta/1681450/buy
XL Anouk 100% viscose NavyMachine-wash
Straight Kurta
http://assets.myntassets.com/v1/assets/images/1681450/2017/1/27/11485516674302-Ano
Navy blue printed
Clothing/Women/Kurtas/Anouk/More
straight kurta,
1099 has a 1099
v-neck,Womenthree-quarter
by Anouk sleeves, curved hem
9fd168b3ce609fdaee8b4c8fe837d1fe
2019-07-28 06:27:51
1849998 +0000
https://www.myntra.com/shrug/global-desi/global-desi-white-net-sheer-shrug/1849998/buy
XS Global Desi80% cotton,White
20% nylonShrug| Hand-washhttp://assets.myntassets.com/v1/assets/images/1849998/2017/4/25/11493122405274-Glo
White net sheer
Clothing/Women/Shrug/Global
shrug with embroidered
719 2399detail,
Desi/More
Womenhas anbyopenGlobal
front,
Desithree-quarter s
fde58f6f7a9b8b76cc30757c840a8817
2019-07-30 07:09:30
8665021 +0000
https://www.myntra.com/sherwani/aj-dezines/aj-dezines-boys-black--grey-sherwani-set/8665021/buy
4-5Y Aj DEZInES 60% Jacquard,
Black40%| Grey
Silk Cotton | Dupatta
http://assets.myntassets.com/v1/assets/images/productimage/2019/1/29/d01afe21-d928-4
Fabric:
BlackVelvet
and grey
Clothing/Boys/Sherwani/Aj
| Machine
sherwaniWash
with
1499
dupatta Black
2999
DEZInES/More
and
Boyswhitebywoven Aj DEZInES
design sherwani, ha
f11d806e761a837ff81369615ebc38e0
2019-08-17 09:37:31
8991019 +0000
https://www.myntra.com/8991019
4-5Y 612 leagueTop fabric: Beige
Pure Cotton
| RedKurta
| Bottom
with Churidar
http://assets.myntassets.com/v1/assets/images/8991019/2019/4/9/0da32239-9a9a-4179-b
fabric: PureRed
Cotton
and beige
| Clothing/Boys/Kurta
Hand-wash
solid kurta with
1047churidar
Sets/6122095
Red
league/More
Boys
and beige straight
by 612 league
knee length kurta, h
87c50d3b515aa726a4917b9ba454beb8
2019-08-11 02:33:26
7217185 +0000
https://www.myntra.com/kurta-sets/jompers/jompers-men-red--beige-self-design-kurta-with-churidar/7217185/buy
L Jompers Top fabric: Beige
Silk Blend
| RedKurta
| Bottomwithfabric:
Churidar
http://assets.myntassets.com/v1/assets/images/7217185/2018/8/24/51335b38-a88e-49b3-
CottonRed
Blend
and|beige
Hand-wash
Clothing/Men/Kurta
self design kurta
1499with
Sets/Jompers/More
churidar
1999 Men Red andbybeige
Jompersstraight knee length ku
04dac287ed030419679b6e3b5f1a8ab2
2019-08-20 01:00:10
9088553 +0000
https://www.myntra.com/dresses/libas/libas-women-red-printed-a-line-dress-with-ethnic-jacket/9088553/buy
M Libas Viscose Rayon
Red| Hand-washEthnic Jackethttp://assets.myntassets.com/v1/assets/images/9088553/2019/3/28/e355672e-a2a8-46ff-a
Red printedClothing/Women/Dresses/Libas/More
woven A-line dress,
989has a round
2199 Women
neck, byshort
Libas
sleeves, asymmetric hem

34
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 5: Conclusion and Future work

Future Work:

• We can improve the accuracy of the model by implementing different model and
make the training dataset much larger.
• Dataset can be improved by adding products of different category. This will help
recommendation of product on different category much more reliable and not
making recommendation on some category weak.
• We can use real time data from the different E-commerce websites so users can
get the real time price and offer on the products.

Conclusion:

In this project we build Fashion products recommender by image search. We tried out
different neural network models for image classification, clustering and different ways to
quantify the similarity between images. We are able to achieve a very great accuracy and
recommend products with similarity score very high. There is under-fitting issue in our
model, which can be one of the things to do in future work by adding more products category
wise. As shown in the Dataset and Features section, though we have a data set of 19000
products due to the limitation on time and machine memory. In the next step, we can try to
train our model on a larger amount of data using batches. This can potentially increase the
accuracy of the model.

35
19IT446, 19IT448, 19IT471, 19IT474 Just Pics

Chapter 6 : References
[1] Santa Clara University IMAGE BASED PRODUCT RECOMMENDATION SYSTEM

[2] S. Bell and K. Bala. Learning visual similarity for product design with convolutional
neural networks. ACM Transactions on Graphics (TOG), 34(4):98, 2015.

[3] T. Deselaers and V. Ferrari. Visual and semantic similarity in imagenet. In Computer
Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on, pages 1777–1784. IEEE,
2011.

[4] A. Dosovitskiy and T. Brox. Generating images with perceptual similarity metrics based
on deep networks. In Advances in Neural Information Processing Systems, pages 658–666,
2016

[5] Image-based Product Recommendation System with Convolutional Neural Networks

[6] Product Recommendation System Using Distance Measure of Product Image Features

[7] Product Recommendation System Video Explanation

36