IRJET - Online Donation based Crowdfunding using Clustering and K-Nearest Neighbour Algorithm

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 07 Issue: 02 | Feb 2020 www.irjet.net p-ISSN: 2395-0072
© 2020, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 3138
ONLINE DONATION BASED CROWDFUNDING USING CLUSTERING AND
K-NEAREST NEIGHBOUR ALGORITHM
G. RAJASEKARAN[1], E. NIROSHA[2], V. SIVARANJANI[3]
1Assitant Professor, Department of Computer Science and Engineering, Jeppiaar SRR Engineering College, Padur
Chennai, Tamil Nadu, India
2,3B.E,Department of Computer Science and Engineering, Jeppiaar SRR Engineering College, Padur, Chennai,
Tamil Nadu, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Crowdfunding is a practice of raising funds from
people to support your project which has bought new life to
charity, i.e, making it easy to donate any amount of money to
help across the globe. Donation-based crowdfunding is the
most preferred mode of fundraising. Crowdfunding through
online platforms knows no boundaries, and has the potential
to travel viral. The problem of high donor attrition i.e, many
donors donate just one occasion or only a few times within a
rather short lifecycle then leave. Thus, it is an urgent task to
analyze the factors of and then further predict the donor
behavior. In this paper, we present a focused study on analysis
of donation recurrence and donor retention to predict the
donor’s interest in donation. Specifically, weproposedamodel,
which has the details of recipient, actual donor and the
verifying person. After the donation process, every donor will
get a proper donation certificateapprovedbythegovernment.
The experimental results will clearly demonstrate the
individual’s interest for donation and to appreciate them to
donate more in their future with a proper secured transaction
with the support of the government.
Key Words: Crowdfunding, Donation recurrence, Donor
retention, Transaction, Attrition.
1. INTRODUCTION
Crowdfunding is an emerging Internet-based fundraising
mechanism soliciting small monetary contributions from
crowd donors to help others in trouble or with dreams [1].
Recent years have witnessed the fastest development of
crowdfunding platforms among which the donation-based
ones are becoming increasingly very popular[1],[2],suchas
Kiva.org 1 [3], and DonorsChoose.org 2 [4]. Leveraging
Internet, crowdfunding has brought new life to charity, i.e.,
making it is very easy to donate any amount of money even
every people to help others across the globe. For example,
Kiva.org is an international nonprofit platform, founded in
2005, with a mission to connect people through lending to
reduce poverty. The accessing donors crowdfund these
projects in increments of $25 or more. Donors may act as
individuals or teams. The critical component for the success
of crowdfunding communities is the recruitment and
continued engagement of donors [4]. However, because of
the non-profit nature, the situation relating to donor
retention for donation-based crowdfunding as well as
traditional charities is extremely serious, i.e., usually, the
donor attrition rate is above 70% [4]. Actually, customer
attrition/churn [5], [6] is crucial and highly focused on in
many commercial scenarios, such as E-commerce, finance
and services. However, for a quite longtime,relevantstudies
on analyzing donor retention in charity have been rather
limited in the literature. Fortunately, with the accumulation
of large-scale user behavior data in crowdfunding platforms,
many data-driven studies which focus on analyzing the user
behaviors have been conducted [7], [8]. For example, Liu, et
al. [7] studied the donation motivation classification in
Kiva.com. Especially, Althoff, et al. [4] penetrate various
factors impacting donorretentioninDonorsChoose.orgfrom
the statistical perspectives which was inspiring for our
research. However, how to comprehensively analyze the
heterogeneous factors affecting and then further predictthe
donor retention or attrition, are still largely unexplored
areas, both in the charity and in other domains. In addition
to these heterogeneous factors,accordingtoourobservation
and analysis, donors’ own behaviors (i.e., donation
recurrence) could particularly reflect their decision on
retention. In fact, donation recurrence prediction is an
unavoidable intermediate goal for forecasting the donor
retention. Thus, in this paper, we try totrack thisproblem by
jointly predicting the donor retention and also the
intermediate goal (predicting donation recurrence).
Although it is necessary to construct the predictions of
donation recurrence and donor retention, as they can alert
platforms that they need to do something before they lose
donors, this is a very challengingtask.First,donorbehaviors,
e.g., donation recurrence, donor retention or attrition, are
influenced by various factors [4], such as their motives and
preferences, their social contacts in crowdfunding
communities, and the characteristicsoftheprojectstowhich
they have recently donated. How to comprehensively
analyze the heterogeneous features and integrate them for
accurate prediction is not a trivial issue.Second,accordingto
our data analysis, the behaviors of donors, especially the
donation recurrence, are highly correlated with their
retention or attrition. How to model the relations of
donation recurrence and donor retention and further
synchronously predict these two behavioral events with a
joint model are quite open problems. Finally,thepresence of
a large amount of censored data [9], [10], i.e., the exact
attrition outcomes of some donors are unobservableorthey
do not perform any behaviors (donation or attrition) during
our monitoring periods, imposes important challenges in

relation to this problem. Becausemanydonorsmaybestill in
the platform in our data and most lost donors do not
explicitly close their accounts when leaving, the censoring
phenomenon is an inescapable concern.
2. RELATIVE WORKS
“A Sequential Approach to Market State Modeling and
Analysis in Online P2P Lending.” Hongke Zhao, Qi Liu,
Hengshu Zhu, Yong Ge, Enhong Chen, Yan Zhu, and Junping
Du,2018.Online peer-to-peer (P2P) advance is an emanate
wealth-management service for individuals, which allows
lenders to directly bid and invest on the listings created by
borrowers without going through any traditional financial
intermediaries. As a nonbank financial platform, online P2P
advance tends to have both high volatility and liquidity.
Therefore, it is of significant importance to comprehend the
hidden market states of the listings(e.g.,hotandcold),which
open venues for intensify business analytics and investment
decision making. However, the problem of market state
modeling remains fetching open due to many technical and
domain provoctions, such as the dynamic and sequential
characteristics of listings. To that end, in this paper, we
present a focused study on market state modeling and
analysis for online P2P advance. Specifically, we first
propose two enhanced sequential models by enlarge the
Bayesian hidden Markov model (BHMM), namely listing-
BHMM (L-BHMM) and listing and marketing-BHMM (LM-
BHMM), for learning the latent semantics between listings’
market states and lenders’ bidding behaviors. Furthermore,
we demonstrate various stimulate applications enabled by
ourmodels, such as bidding prediction and herding
detection. Finally, we construct substantial experiments on
two real-world data sets and make some deep analysis on
bidding behaviors, which clearly validate the potency of our
models in terms of different applications and also disclose
some interesting business findings.
3. PROPOSED SYSTEM
In this work, the client and the donor have to fill their
personal details which will be verified by the third party,the
verifying agent appointed by the government. The verifying
agent will accept the details proceeds secured transaction
from the donors to the clients. This system uses the
clustering algorithm to filter the data from large scale of
datasets and uses K-Nearest neighbor algorithm for
clustering, which will cluster the similar data items from
large dataset according to the user preference. This system
will automatically notifies the donors on any particular day,
for example, on their birthday, and appreciate the donors to
further improve their sequence of donation.Thissystem will
make a secure transaction from the donors to the clients
only after verifying whether the donors and the clients are
authorized persons. And alsothiswill improvethe eagerness
of donors for donation.
4. SYSTEM DESIGN
The recipient has to fill the registration form with required
mandatory input fields and the personal details completely.
These details will get further verified by the government
authority or the verifying agent appointed by the
government. The donor has to fill the registration form with
required mandatory input fields and the personal details
completely. These details will get further verified by the
government authority or the verifying agent appointed by
the government. By using the token, generated after the
completion of verifying recipient’sdetailsbythegovernment
authority. The recipient will log in only suing the unique
token to request money. Their requestcontainsthedetailsof
the category for which the recipient needs money.
Figure 1: System Architecture

The recipient can view their request details and status of
donation for their requests. By using the token, generated
after the completion of verifying donor’s details by the
government authority. The donor will log in only using the
unique token to donate money. Their request contains the
details or the category for which the donor is willing to
donate. The donor can view the history of their donations.
After verifying all the details of donor and the recipient, the
verifying person will approve the details of recipient and the
donor and further allows the process of funding. After the
funding process, the donor will be provided a certificate by
the government. The donor will get notifications
automatically to appreciate their donation funding.
USECASE DIAGRAM
Unified Modeling Language (UML) may be a systematized
general-purpose modeling language within the field of
software engineering. The standard is managed and was
created by the thing Management Group. UML includes a
group of graphic notation techniques to make visual models
of software intensive systems. This language is employed to
specify, imagine, modify, build and document the artifactsof
an object oriented software intensive system under
development.
Use case diagram consists of two parts:
Use case: A use case describes a sequence of actions that
provided something of quantifiable value to an actor and is
drawn as a horizontal ellipse.
Actor: An actor could also be an individual , organization or
external system that plays a task in one or more interaction
with the system.
CLASS DIAGRAM
A Class diagram shows how the dissimilar entities
interconnected to each other in the Unified Modeling
Language was a type of static structure diagram that
illustrate the structure of a system by demonstration the
system's classes, their attributes, operations (or methods),
and the relationships among objects.
COLLABORATION DIAGRAM
UML Collaboration Diagrams illustrate the link and
interaction between software objects. They necessitate use
cases, system operation contracts and domain model to
already exist. The collaboration diagram embellished
messages being sent between classes and objects.
CLASS DIAGRAM
A Class diagram shows how the dissimilar entities
interconnected to each other in the Unified Modeling
Language was a type of static structure diagram that
illustrate the structure of a system by demonstration the
system's classes, their attributes, operations (or methods),
and the relationships among objects.
SEQUENCE DIAGRAM
A Sequence diagram is a kind of interaction diagram that
shows how procedure manage with one anotherandinwhat
order. It is a build of Message Sequence diagrams are
sometimes called event diagrams, event scenarios and
timing diagram.
ACTIVITY DIAGRAM
Activity diagram is a graphical representation of workflows
of gradual activities and actions with support for
possibility,looping and consistency.
The most important shape types:
• Rounded rectangles represent activities.
• Diamonds represent decisions.
• Bars represent the start or end of consistent activities.
• A black circle represents the start of the workflow.
• An encircled circle represents the end of the workflow.
DATA FLOW DIAGRAM
The Data Flow Diagram is a graphical representation of the
“flow” of data through an information system, modeling its
point. It is a preliminary step used to create an overview of
the system which can later be elaboratedData FlowDiagram
can also be used for visualization of data processing.
5. MODULES
The system module is categorized intofourmodulesnamely,
1. Recipient Authentication
2. Donor Authentication
3. Recipient’s Requests
4. Donor’s Donation Funding
I. RECIPIENT AUTHENTICATION
The recipient has to fill the registration form with required
These details will get verified by the government authority
or the verifying agent appointed by the government as same
as the donor’s verification. After successful verification, the
recipient will get the notifications regarding their donation
with securedauthentication. Therecipient’schatbotcontains
the details or the category for which the recipient needs
donation.

II. DONOR AUTHENTICATION
The donor has to fill the registration form with required
These details will get verified by the government authority
or the verifying agent appointed by the government. After
successful verification, the donor will get the registration
and the authenticated mail.
III. RECIPIENT’S REQUESTS
By using the token, generated after the verification of
recipient’s details by the government authority. The
recipient will log in only using the unique token to request
money. The request contains the details of the category for
which the recipient needs money. Recipient can view their
request details and status of donation for their requests.
IV. DONOR’S DONATION FUNDING
After verifying all the details of donor and the recipient, the
verifying person will approve the details of recipientand the
donor and further allows the process of funding. The donor
will be displayed with the recipient details in the category
for which the donor is willing to donate. The donor will
select the recipient and further donates. After the funding
process, the donor will be provided by a certificate by the
government.
6. CONCLUSION
We presented a focused study on prospecting the donation
careers in crowdfunding. By collecting and analyzing large-
scale real-world data, we specifically formalized predicting
tasks by automaticallygeneratingnotificationstothedonors,
in order to appreciate their process of donation funding.
Then, using a clustering algorithm and K-Nearest neighbor
algorithm, we proposed a model which could integrate
heterogeneous features to jointly model the donor activities
in predicting their donation interests. Additionally, we
designed multiple innovative constraints and incorporated
them into objective functions for modeling the censoring
phenomenon and dependence relations of different
behaviors. In experiments, we analyzed the donations in
crowdfunding and validated the prediction performances
from various aspects. The experimental results clearly
demonstrated the effectiveness of our proposed models for
analyzing and predicting the donors and appreciating them
for further more donations with a trustworthy funding.
7. FUTURE ENHANCEMENT
Our study may bring some new insights fromtheapplication
view of crowdfunding and the technical view of exploiting
deep learning for survival analysis to the research
communities. In the future, we will apply and improve our
models for other scenarios, such as traditional charity
activities, especially applied to survival data with modeling
collaborative tasks in some other domains, such as a device
failure in engineering, predicting student dropout, and
prospecting the career development.
VIII. REFERENCES
[1] H. Zhao, Y. Ge, Q. Liu, G. Wang, E. Chen, andH.Zhang, “P2p
lending survey: Platforms, recent advances and prospects,”
ACM Transactions on Intelligent Systems and Technology
(TIST), vol. 8, no. 6, p. 72, 2017.
[2] H. Zhao, H. Zhang, Y. Ge, Q. Liu, E. Chen, H. Li, and L. Wu,
“Tracking the dynamics in crowdfunding,” in Proceedings of
the 23rd ACM SIGKDD International Conference on
Knowledge Discovery and Data Mining. ACM, 2017,pp.625–
634.
[3] J. Choo, D. Lee, B. Dilkina, H. Zha, and H. Park, “To gather
together for a better world: Understanding and leveraging
communities in micro-lending recommendation,” in
Proceedings of the 23rd international conference on World
wide web. ACM, 2014, pp. 249–260.
[4] T. Althoff and J. Leskovec, “Donor retention in online
crowdfunding communities: A case study of donorschoose.
org,” in Proceedings of the 24th International Conferenceon
World Wide Web. International World Wide Web
Conferences Steering Committee, 2015, pp. 34–44.
[5] L. J. Rosenberg and J. A. Czepiel, “A marketing approach
for on Knowledge and Data Engineering, vol. 28, no. 4, pp.
901–911, 2016. [12] G. Li, Y. Zheng, J. Fan, J. Wang, and R.
Cheng, “Crowdsourced data management: Overview and
challenges,” in Proceedings of the 2017 ACM International
Conference on Management of Data. ACM, 2017, pp. 1711–
1716.
[13] A. I. Chittilappilly, L. Chen, and S. Amer-Yahia, “A survey
of general-purpose crowdsourcing techniques,” IEEE
Transactions on Knowledge and Data Engineering, vol. 28,
no. 9, pp. 2246–2266, 2016.
[14] G. Li, J. Wang, Y. Zheng, and M. J. Franklin,
“Crowdsourced data management: A survey,” IEEE
Transactions on Knowledge and Data Engineering, vol. 28,
no. 9, pp. 2296–2319, 2016.
[15] G. Li, J. Wang, Y. Zheng, J. Fan, and M. J. Franklin,
Crowdsourced Data Management: Hybrid Human-Machine
Data Management. Springer, 2018.
[16] Y. Zheng, R. Cheng, S. Maniu, and L. Mo, “On optimality
of jury selection in crowdsourcing,” in Proceedings of the
18th International Conference on Extending Database
Technology, EDBT 2015. Open-Proceedings. org., 2015.

IRJET - Online Donation based Crowdfunding using Clustering and K-Nearest Neighbour Algorithm

More Related Content

Similar to IRJET - Online Donation based Crowdfunding using Clustering and K-Nearest Neighbour Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET - Online Donation based Crowdfunding using Clustering and K-Nearest Neighbour Algorithm