You are on page 1of 5

Proceedings of the Second International Conference on Electronics and Renewable Systems (ICEARS-2023).

IEEE Xplore Part Number: CFP23AV8-ART; ISBN: 979-8-3503-4664-0

Property Price Prediction, Litigation Detection and


Prevention using Machine Learning
2023 Second International Conference on Electronics and Renewable Systems (ICEARS) | 979-8-3503-4664-0/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICEARS56392.2023.10085347

Geetha D. Devenagavi Bukkapatnam Mohammed Aleem C Saiswaroopa


School of Computing and Information School of Computing and Information School of Computing and Information
Technology Technology Technology
REVA University REVA University REVA University
Bengaluru, India Bengaluru, India Bengaluru, India
dgeetha@reva.edu.in mohammedaleem1404@gmail.com csaiswaroopasaiswaroopa@gmail.com

Jonnalagadda Karthik Niharika Chinnappannagari


School of Computing and Information School of Computing and Information
Technology Technology
REVA University REVA University
Bengaluru, India Bengaluru, India
karthikjonnalagadda032@gmail.com niharika02072001@gmail.com

Abstract— The value of any property will gradually provides the design flow and major concepts of the
increase day to day according to the demand, popularity, problem and the built model, section 3 illustrates about the
and many other individual features. As per the report of the workflow of the model, section 4 gives the final idea and
Indian Ministry of Law & Justice, the number of property usage of the model and section 5 conclude the paper.
fraudulent cases have been increasing gradually. Nearly 12-
13% of the civil cases are increasing every year. Now a days II. DESIGN FLOW
many managers and real estate dealers are committing
frauds over the prices and the value of a particular property.
The design flow of this model will be done with respect to
To avoid this problem, a computing model is designed which the following objectives:
computes the price of a particular property according to the 1.Create a database to store all the necessary information
previous price ranges, etc. Barcodes of different colors are and the documents of all the property.
used to classify the different types of ownerships of the 2.Assign a barcode according to the survey number of the
property for its survey numbers which are considered as the property.
unique ID number for a property assigned after its first 3.Manage the access to different officials and users.
registration by authorized governance. 4.Validation of information every month and report
should be updated.
This model helps to predict estimated price of property
for a particular time being. This model helps us to avoid the To fulfill the first objective, creating a blue print
fraudulent people though the customer does not have any of a database, a schema which will help us to manage the
direct contact with the property right now. The litigation relationship among the entities. Duplicate data base is also
detection will be the major part of the model where system created to back up all the information which helps to
will detect the ongoing litigation count over a property. prevent the data loss as well as security. After creating the
database, a web application is created which will be the
Keywords— Property price, Survey number, Barcode, most essential part of the model through which all the
Property litigation. users can access and experience the model. The access
will be managed only to the officials who undertakes all
I. INTRODUCTION
the activities in the department of law and the revenue
Enquiry about a property before purchasing it by a department. All the document uploaded will be stored to
transferee is the major aspect of the decision about its the least possible quality for printing in order to reduce the
purchase. A transferee should be aware of all the storage usage as well as the fetching time.
information about the property like its dimensions and The second objective links the assigning of
fortunes. When it comes to the transferor, they can barcode to a survey number, which is said to be the unique
manipulate the property documents and fake their
number given to a property. Different colors of barcodes
customer. Whereas some will pretend themselves as the
owners of the estates and tends to transfer it to the is assigned to different of ownership. For example, usual
customer who is not aware of the ownership and the citizen owned property will be assigned with black and
property fortunes. Past data of a property and its white barcode, the government owned property will be
documents, assigned to a unique identification of a citizen assigned with green colored barcode, bank owned
which is stored by revenue department is used for this property will be assigned with blue colored barcode, any
model through central database . This computing model seized property under some cases will be assigned with
which will compute the price of a particular property red color.
according to the previous price ranges, etc. This model will In the third objective access to the officials and
use some inbuilt statistical methods and function in python users is managed, so that law department can access only
libraries to build this model. In addition to the price certain part of the application which consists only the case
prediction, model also detect the litigations present if any, related issues and all the updates done by their
which helps to decrease the number of civil cases in the department, similarly revenue department is also assigned
world. This paper is organized as follows. Section 2 only certain part of the application where they also can

979-8-3503-4664-0/23/$31.00 ©2023 IEEE 1301

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on April 18,2023 at 06:47:55 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Electronics and Renewable Systems (ICEARS-2023).
IEEE Xplore Part Number: CFP23AV8-ART; ISBN: 979-8-3503-4664-0

access only the part of application which consists only the in the courts[13]. To reduce these problems many
registration related information and the updates done by governments are implementing several schemes and
their respective department. When it comes to the processes. This model can be efficient among them
legislature department access, they can observe whole and easy to use as well. The ownership of the property
access combined of both revenue department as well as will be identified easily using the color of the barcode
the department of law. If any change is required to the assigned to its document. The current status of their
type of ownership, two level legislature permissions is civil cases will be identified easily if any. If any
required i.e., Sub level officer and the regional level pending cases present, then the property cannot be
officer. The change of ownership request will be sent sold to anyone and its transfer of ownership is
directly to the regional level officer, who will redirect the impossible. Only the legislature officer has the right to
same to the sub level officer to raise and enquire a query allow the registration process of pending case property
about the survey number. When it comes to end user as a special case if required upon the request of both
access (citizen level access), they can have access to the the transferer and the transferee.
properties they own and their related documents. They are The price of every property will change time to
supposed to login with the unique identification number time, in most of the cases it increases and in very least
provided by the country to track which all the properties number of cases it will decreases due to unfortunate
owned by them. The banks also have access to the reasons and dependencies [4]. Every individual need
application with limited terms of usage. Ultimately this to be aware of the cost of their property. Many people
database can also be accessed by some other departments will own a land at a different place and they will reside
like income tax department, police department, at other place or else in a different country. They
Commissionerate etc. for their respective back-end might not aware of the current estimate price of their
verifications. property [10][8]. If they want to sell it, many people
In the final objective, requests for validation of will approach brokers present at that local place. There
documents and case reports can be made and the is a chance that they can help the transferee to buy at
verification report and updating it shall take place every lower cost than the actual and get bribe in return.
month which can reduce the manipulation of records. Some leaders in the illiterate area can also occupy
them if the owner of the property is not actively taking
care of the property. Many cases of similar kind are
filing every day and increasing every year all over the
worlds, especially in the countries like India, Pakistan,
Afghanistan, Iran, African countries, and many
undeveloped countries.
IV. METHODOLOGY
A. Property price prediction
While calculating the assumed price of a property,
system will use statistical methods like Mean, Standard
Deviation (SD), Slope etc. The prices of surrounding
properties are considered using clustering method of
machine learning, which will help to predict the price of
actual property according to the property location. The
time complexity of this model is better than the previously
developed existing models. To calculate the assumed price
of the property, the history of property price of previous
sale along with that year of sale is needed. The calculation
of the price prediction will be followed as:
Mean is the average of the given numbers and is
calculated by dividing the sum of given numbers by the
total number of numbers. Mean = (Sum of all the
observations/Total number of observations).

∑𝑥
Mean, μ = 𝑛
.
Fig. 1. ER diagram of the model database. Where,
Σx = sum of all the observations recorded,
III. LITERATURE SURVEY n = number of observations recorded.
The main problem of land colonizing, document
manipulation and all other illegal activities takes place Standard Deviation is a measure which shows how
due to the lack of knowledge about their properties much variation (such as spread, dispersion, spread,) from
and their fortunes. Even false documents submission the mean exists. The standard deviation indicates a
to the transferee also been a large scale filed case type “typical” deviation from the mean. It is a popular measure

979-8-3503-4664-0/23/$31.00 ©2023 IEEE 1302

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on April 18,2023 at 06:47:55 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Electronics and Renewable Systems (ICEARS-2023).
IEEE Xplore Part Number: CFP23AV8-ART; ISBN: 979-8-3503-4664-0

of variability because it returns to the original units of previous steps to the price of previous sale of the
measure of the data set. Like the variance, if the data property.
points are close to the mean, there is a small variation
whereas the data points are highly spread out from the
mean, then it has a high variance. Standard deviation
calculates the extent to which the values differ from the B. Litigation detection and prevention
average. Standard Deviation, the most widely used
measure of dispersion, is based on all values. Therefore, a
change in even one value affects the value of standard In the litigation detection part of the model, the
deviation. It is independent of origin but not of scale. It is governing authority are going to create a central database
also useful in certain advanced statistical problems. which is maintained by the government. In order to
prevent data loss and ensure security, a replica database
(𝑥𝑖−𝜇)2 that stores all the information is maintained. The most
Standard Deviation, σ = √
𝑁 crucial component of the model will be the web
Where, application, which will be developed after the database,
and will allow all users to access and interact with the
μ = mean of the data, model. Access will only be controlled for the employees
xi = ith observation in the recording, that handle all operations for the revenue and law
departments. To save on storage space and retrieval time,
N = number of observations. all uploaded documents will be saved at the lowest quality
feasible for printing.

Slope describes the steepness of a line. The slope of After creating the database, the registration authority
any line remains constant along the line. The slope can needs to issue the barcodes for the property documents
also tell you information about the direction of the line on according to the ownership of the property. Various
the coordinate plane. Slope can be calculated either by barcode colors are assigned to different types of
looking at the graph of a line or by using the coordinates of ownerships exists. For instance, the typical citizen's
any two points on a line. property will be assigned a black and white barcode, the
government's property will be assigned a green barcode,
the bank's property will be awarded a blue barcode, and
any property that has been taken in specific circumstances
will be assigned a red hue. When updating the records and
the issue of the barcodes is completed, the model
development will move on to the access control part of
the application.
Access to the officials and users is controlled so that
the law department can only access a specific part of the
application that only contains case-related issues and all
updates made by their department. In a similar manner,
Fig. 2. Slope of a line the revenue department will be assigned a specific part of
the application and be able to access only the part of the
𝑦 −𝑦
Slope, (x1, x2, y1, y2), m = 𝑥2−𝑥1 . application that only contains registration-related
2 1
information and updates made by their respective
Where,
departments.
x1, y1 are coordinates of previous recording,
The revenue department as well as the department of
x2, y2 are the coordinates of the newest
law can both be seen when it comes to the access to the
recording.
legislature's departments. If the type of ownership needs
Slope of a line is used to find the rate of change in the to change, two levels of legislative approval the regional
data points/ coordinates, in this model use slope to find level officer and the sublevel officer are needed.
out how the rate of change took place in the price of a The change of ownership request will be forwarded to
property using history of its sales and the year of sale. the sublevel official to raise and enquire a question
After the calculating the slope of entries system will regarding the survey number after being sent straight to
calculate the standard deviation among the elements of that the regional level officer. End users (citizen level access)
array. Finally, the current price will be calculated using the can have access to the properties they own and the
below formula documents that are related to those properties. To track all
the properties they possess, they must log in using the
Current price = Price of previous sale + [(current year – special identification number supplied by the nation. The
year of previous sale) x application is also accessible to banks under certain
SD]. conditions. In the end, other agencies like the income tax
department, police department, Commissionerate, etc. can
The final price will be calculated by adding the values also use this database for their individual back-end
of product of the year gap between last sale and present verifications.
year as well as the standard deviation recorded in the

979-8-3503-4664-0/23/$31.00 ©2023 IEEE 1303

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on April 18,2023 at 06:47:55 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Electronics and Renewable Systems (ICEARS-2023).
IEEE Xplore Part Number: CFP23AV8-ART; ISBN: 979-8-3503-4664-0

To achieve the successful implementation of all the than the 2-dimensional codes. And the persons with
methods proposed, the time to time updating of records to minimal technical knowledge also learn the control of
the central database is much necessary, where all the data current code reading system.
of the properties will need to be digitalized in the whole
V. CHALLENGES/ISSUES FACED
country along with the registration history details as old as
available in the old records. The assigning of survey  The unusual aspect in the society like pandemics,
numbers to the newly registered property to the database natural disasters, etc., can make the model
will be updated along with the old documents of the same inaccurate in the terms of price prediction.
and the parent documents of the property if the partitions  Time to time data updating from both the revenue
are done. The partition details also need to be in the department as well as the law department.
records of the database to prevent any property size
related litigations.  Access to the data which is not updated online (not
present in the authorized servers. For example, old
This model also helpful to identify the properties records of properties past digitalization).
owned by an individual and it is easy to track the income  Implementing this model in remote area, which
source and the net worth of the individual. does not have proper access to the internet.
VI. CONCLUSION
This paper proposes a novel technique of registration
process to the properties with added security, where the
society can reduce the number of civil cases increasing
every day. It also provides access to the properties owned
by every individual and to the governing body. The model
and its working are explained very clearly along with the
kind of access that different bodies have. The proposed
model will help the government to identify the illegal
activities which takes place in several departments and
keep track of them very easily. This model also helps for
banking sector while processing the loan applications. The
future studies include the addition of few more features
according to the usage.
Finally, the litigation detection achieved using the
newly proposed registration process where the application
will not be processed if any pending cases exists on a
property.

Fig. 3. Flow chart of a property registration Expected number of civil cases


process after implementing this model. for each 5 years
C. Barcodes 20000000
A barcode, consisting of bars and spaces, is a machine- 10000000
readable representation of numerals and characters. It is
widely used for billing purposes in all the shopping areas. 0
It consists of the product name, ID (if applicable), price 2015 2020 2025 2030
and other details if required. A 1-dimensional barcode is
very easy to read and it can store up to 85 characters, Litigations
whereas the 2-dimensional barcodes like QR codes can
store up to 7089 characters. Fig. 4. Calculation assumption about number
When the registration of a property completes, a of civil cases after implementation of this
barcode sticker will be sticked to the original document of model.
registration which is of the colour according to the
ownership of the property. The information will be
updated time to time to the database about that particular The number of civil cases is assumed to
property and keep the user updating for every change. If be decreased after implementing this novel
any change is made unaware of the user, then he/she can model of a property registration. There will be
report to the higher officials about this concern and clarify
their queries. If a property is divided as a process of possible decrement of 10% civil cases every
partition, the history of all the partitions also be recorded to year, the number of civil cases will be decreased
the database. Using QR codes also be a good option but, as assumed if this model is implemented in the
the reading and accessing of 1-dimensional codes is easier year of 2025 as mentioned in fig.4. Addition to

979-8-3503-4664-0/23/$31.00 ©2023 IEEE 1304

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on April 18,2023 at 06:47:55 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Second International Conference on Electronics and Renewable Systems (ICEARS-2023).
IEEE Xplore Part Number: CFP23AV8-ART; ISBN: 979-8-3503-4664-0

this the number of corruption cases also [14] Yiyun Chen, “Exploration of the regional correlation and network
structure characteristics of land prices”, Wuhan University, China,
decreases with added features to the application. 04 January 2023.
[15] Paul Waddell, Gudmundur F. Ulfarsson, “Dynamic Simulation of
Real Estate Development and Land Prices within an Integrated
ACKNOWLEDGEMENT Land Use and Transportation Model System”, University of
Washington, Box 353055, Seattle, January 2003.
The authors are grateful to REVA University, School
of Computing and Information Technology, Bengaluru, for
providing the facility to carry out advanced research work
on the premises of the university.
REFERENCES

[1] Aniket Singh, Adarsh Kumar Singh, Aditya Raj, Harshit Jain Asha
M S, “House Price Prediction Analysis using Machine Learning”,
www.ijcrt.org, Department of Computer Science and Engineering,
Dayananda Sagar Academy of Technology and Management
(DSATM), Bengaluru, Karnataka, India, 4 April 2022.
[2] Jungsun Kim, Jaewoong Won, Hyeongsoon Kim, Joonghyeok Heo,
“Machine-Learning-Based Prediction of Land Prices in Seoul”,
https://www.mdpi.com/2071-1050/13/23/13088, Kyung Hee
University, Seoul 02447, Korea, 26 November 2021.
[3] John Ade-Ojo, “Predicting House Prices with Machine Learning”,
https://towardsdatascience.com/predicting-house-prices-with-
machine-learning-62d5bcd0d68f, 17 January 2021.
[4] Shubham Singh, Monika Nag K J, “Land Price Prediction using
Machine Learning Algorithm”,
https://www.irjet.net/archives/V7/i5/IRJET-V7I5382.pdf, Dept. of
Information Science and Engineering, NIE, Mysore, India, 5 May
2020.
[5] Daniel Di Benedetto, Leimin Gao, Yiwei Huang, Neha Sharma,
Dongying Wang, “HOUSING PRICE PROJECT REPORT”, 14
July 2018.
[6] April Taylor, “Median Housing Price Prediction Model”, D. M.
Pan National Real Estate Company, Southern New Hampshire
University, Southern New Hampshire University, United States, 18
January 2020.
[7] Ujjwal Tarika, Shubham Singh, “House Price Prediction”, School
of Computing Science and Engineering Department of Computer
Science and Engineering Galgotias University, Greater Noida,
India, December 2021.
[8] Neha Soni, Poonam kanuajiya, Rahul Dubey, “Design and
Development of House Price Prediction”, Sagar Institute of
Science and Technology (SISTec), Gandhinagar, Bhopal, India, 23
May 2021.
[9] Sumanth Mysore, Abhinay Muthineni, Vaishnavi Nandikandi,
Sudersan Behera, “Prediction of House Prices Using Machine
Learning”, https://www.ijraset.com/best-journal/prediction-of-
house-prices-using-machine-learning, Department of Electronics
and Computer Engineering, Sreenidhi Institute of Science and
Technology, Hyderabad, Telangana, India, June 2022.
[10] Amey Thakur, Mega Satish, “BANGALORE HOUSE PRICE
PREDICTION”, Department of Computer Engineering, University
of Mumbai, Mumbai, MH, India,
https://www.academia.edu/79097924/BANGALORE_HOUSE_PR
ICE_PREDICTION?email_work_card=view-paper, 9 September
2021.
[11] Tarun Jain, “ Examining the Concept of 'Cause of Action' in India”,
Supreme Court of India, New Delhi, India,
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1087352, 30
January 2008.
[12] Chantraale price, “ Civil Case types”,
https://study.com/learn/lesson/civil-case-types-examples-civil-
court.html, Georgia state university, Atlanta, United States, 2
August 2022.
[13] Bimali Y.M.Y, Rodrigo U.S.D, Thenuka Dharmaseelan, “A
Machine Learning-Based Solution for Predicting Land Values”, ,
Department of Software Engineering, Faculty of Computing, Sri
Lanka Institute of Information Technology (SLIIT). Malabe, Sri
Lanka, June 2021.

979-8-3503-4664-0/23/$31.00 ©2023 IEEE 1305

Authorized licensed use limited to: REVA UNIVERSITY. Downloaded on April 18,2023 at 06:47:55 UTC from IEEE Xplore. Restrictions apply.

You might also like