MICRO AND NANO ROBOT

BASED DRUG DELIVERY SYSTEM

A Seminar Report

Submitted by

ABHIMANYU S

REG NO: 2101040095

FIFTH SEMESTER

ELECTRONICS ENGINEERING

DEPARTMENT OF ELECTRONICS ENGINEERING

SREE NARAYANA POLYTECHNIC COLLEGE

KOTTIYAM, KOLLAM.

2023 – 24
 MICRO AND NANO ROBOT

BASED DRUG DELIVERY SYSTEM

A Seminar Report

Submitted by

ABHIMANYU S

REG NO: 2101040095

FIFTH SEMESTER

ELECTRONICS ENGINEERING

DEPARTMENT OF ELECTRONICS ENGINEERING

SREE NARAYANA POLYTECHNIC COLLEGE

KOTTIYAM, KOLLAM.

2023– 24
 MICRO AND NANO ROBOT

BASED DRUG DELIVERY SYSTEM

A Seminar Report

Submitted in partial fulfilment of

The requirement for the award of

Diploma in Electronics Engineering

Of the board of technical engineering education

Submitted by

ABHIMANYU S

REG NO: 2101040095

FIFTH SEMESTER

ELECTRONICS ENGINEERING

DEPARTMENT OF ELECTRONICS ENGINEERING

SREE NARAYANA POLYTECHNIC COLLEGE

KOTTIYAM, KOLLAM.
 DEPARTMENT OF ELECTRONICS ENGINEERING

SREE NARAYANA POLYTECHNIC COLLEGE

KOTTIYAM, KOLLAM.

2023 – 24

CERTIFICATE

This is to certify this report entitles “micro and nanorobot based

drug delivery system” is a Bonafede record of the seminar presented by
ABHIMANYU S of 5th semester diploma under our guidance towards the
partial fulfilment of the requirement for the award of Diploma in Electronics
Engineering of the Board of Technical Engineering Education of Kerala during
the year 2023– 2024.

RAKNAS SANKAR RAKNAS SANKAR

Head of the Department Lecturer

Electronics Engineering Electronics Engineering

 ACKNOWLEDGEMENT

I would like to express my sincere gratitude and heartful indebtness to my Professors,

Department of Electronics Engineering for their valuable guidance and encouragement in
pursuing this seminar. I am thankful to Mr RAKANS SANKAR, Head of the Department and
also my seminar guide of Department of Electronics Engineering for their help and support. I
also acknowledge other members of faculty in the Department of Electronics Engineering and
all my friends for their whole-hearted cooperation and encouragement.

Above all I am thankful to the almighty.

ABHIMANYU S

6
 PREFACE

I have made this report file on the topic “micro and Nano robot based Drug delivery
system”; I have tried my best to elucidate all the relevant detail to the topic to be included in
the report. While in the beginning I have tried to give a general view about this topic.

My efforts and wholehearted co-corporation of each and every one has ended on a successful
note. I express my sincere gratitude my guide Mr RAKNAS SANKAR SIR who assisting me
throughout the preparation of this topic. I thank him for providing me the reinforcement,
confidence and most importantly the track for the topic whenever I needed it.

7
 CONTENTS

TOPIC PAGE NO

ACKNOWLEDGEMENT i

PREFACE ii

CONTENTS iii

LIST OF FIGURES vi

LIST OF TABLES vii

ABSTRACT 1

CHAPTER 1 INTRODUCTION 2

CHAPTER 2 CLASSIFICATION OF ROBOTS 4

CHAPTER 3 NANO ROBOTS

CHAPTER 4 MICOR ROBOTS

CHAPTER 5 STEPS IN DRUG DELIVERY

10

CHAPTER 6 STRUCTURE OF NANO ROBOTS USED IN DRUG DELIVERY

11

CHAPTER 7 NANO ROBOTS INSIDE A BLOOD VESSEL 12

CHAPTER 8 ADVANTAGES

13

CHAPTER 9 DISADVANTAGES

14

CHAPTER 10 APPLICATIONS

8
16

CHAPTER 11 DIFFERENT TYPES OF NANO ROBOTS USED IN MEDICAL FIELD

18

CHAPTER 12 CONCLUSION 19

9
 LIST OF FIGURES

FIGURE PAGE NO

Fig : STRUCTURE OF NANO ROBO USED IN DRUG DELIVERY 11

Fig : NANO ROBOTS INSIDE A BLOOD VESSEL 12

Fig : TYPES OF NANO ROBOTS USED IN MEDICAL FIELD 18

10
 ABSTRACT

Progress in the drug delivery system in the last few decades has led to many advancements for
efficient Drug delivery. Both micro and Nano robots, are regarded as superior drug delivery

11
systems to deliver drugs Efficiently by altering other forms of energy into propulsion and
movements. Furthermore, it can be Advantageous as it is directed to targeted sites beneath
physiological environments and conditions. They Have been validated to possess the capability
to encapsulate, transport, and supply therapeutic contents Directly to the disease sites, thus
enhancing the therapeutic efficiency and decreasing systemic side effects Of the toxic drugs.
This review discusses about the micro and Nano robots for the diagnostics and management of
diseases, types of micro, and Nano robots, role of robots in drug delivery, and its biomedical
Applications.

12
 CHAPTER 1
INTRODUCTION

The robot systems have significantly improved human beings ‘capabilities in detecting,
interrelating, operating, and altering different systems worldwide. The convergence of various
technologies has revolutionized different fields, especially robotic technologies for therapeutic
purposes to enhance medical care. Industrial robots were mainly established to mechanize
predictable and hazardous macroscale developed liabilities, while the therapeutic robot devices
are synthesized exclusively for the diagnostics and management of diseases. Hence, contrary
to conventional (old) robots, which are fictional through massive mechanical systems, the
therapeutic robots require decreased sections and insolent resources for complex and detailed
procedures as well as improve reproducibility in the human body. Impressive progress has been
made in medical robotics through technological advancements, such as combination in control
theory, motors, constituents, and medical imaging, which can be observed from the rise in
surgeon-patient acceptance . Like, surgical robot systems (da Vinci system) permit conversion
of surgeon’s hand action into minor, accurate actions of small devices inside the patient’s body.
Although robotic systems are prevalent for slightly invasive surgery, there are still leading
mechanical problems and tasks . Robots have emerged as a favorable option for the targeted
and controlled delivery of drugs. The main benefit of robotic devices is their controlled, precise,
and site-specific drug delivery capability, compared to other conventional therapeutic delivery
systems of smaller sizes. Robots like stimuli-responsive and surface-functionalized magnetic
helical micro swimmers have not only the capability to target particular single cells but also
the desired areas of the tissues for the controlled and targeted delivery of cells, genes, or drugs
. Commonly, the mechanism of cargo/ drug release in a small robot-like (micro/Nano robots)
is initiated not actively through limited modifications in physiological environments (such as
temperature, pH) at the target site. However, the delivery of promising off-target therapy may
be produced through complex physiological environments or unpredicted local variations.
Hence, establishing an assimilated mobile robot (micro/ Nano robot) that may be superficially
activated will deliver an adequate quantity which is further manageable and satisfactory.
Preferably, a therapeutic delivery platform must exploit an equal exterior power source
autonomously and mutually for movement and on-demand activated therapeutics. Still, the

13
interaction between these two tasks is exciting while consuming the equivalent amount of
energy .The rapid advancement of materials science, molecular biology, mechanical dynamics,
artificial intelligence, and other disciplines accelerates the emergence of new technologies.
Micro/Nano robots increasingly entered the minds of scientists. Richard Feynman suggested
the use of micro robots in medical therapy for the first time in 1959. The term ‘Nano robot’
was coined at that point. Micro/Nano robots have been extensively utilized in the medical
sector as a growing technology, with tasks such as auxiliary operation, medical diagnostics,
and drug/cells delivery. In terms of drug delivery, unlike the conventional approach, which
depends on blood flow to reach the target, micro/Nano robots may accomplish autonomous
mobility, allowing us to distribute controlled nanoparticles to challenging regions . Recent
advances in micro and Nano robots, classification, and their usage for drug delivery, targeted
drug delivery, and other medicinal applications are addressed in this article.

14
 CHAPTER 2

CLASSIFICATION OF ROBOTS

15
Robots can be classified based on their functionality and design into various categories:
Industrial Robots: Designed for manufacturing tasks, such as assembly, welding, or painting
in factories. Service Robots: Intended for non-manufacturing roles, including cleaning robots,
medical robots, and personal assistant robots. Medical Robots: Used for medical procedures,
surgery, or rehabilitation, enhancing precision and minimizing invasiveness. Autonomous
Robots: Capable of operating without direct human control, often employed in fields like
agriculture, exploration, or transportation. Humanoid Robots: Resemble humans in appearance
and behavior, aimed at tasks that benefit from human-like interactions. Aerial Robots (Drones):
Unmanned aerial vehicles used for tasks such as surveillance, delivery, or inspection. Mobile
Robots: Equipped with wheels or legs for movement, commonly found in logistics, security,
or exploration. Educational Robots: Designed for educational purposes, helping users learn
programming, robotics, or other STEM concepts. Entertainment Robots: Created for
amusement, including robotic toys, companions, or performers. Telepresence Robots: Enable
users to remotely interact with distant environments, often used for business meetings or
healthcare consultations. These categories highlight the diverse applications and forms of
robots in various industries and daily life.

CHAPTER 3

MICROROBOTS

16
Micro robots have demonstrated a great potential to perform various important functions, for
example, delivery of the drugs, Manipulation of the cells, micro assembly, and bio sensing
through Manual mechanisms [8]. For example, targeted delivery was established by exhausting
magneto tactic bacteria beneath DC magnetic Field gradients [9]. Many researchers have
shown micro robots to Transport target substances like cells and chemicals by employing
Magnetic fields. Moreover, micro grippers have been established For micro robots using a
unique system known as micro-electromechanical systems, a technology that may be employed
to Expand micro robots functionality. The kinematic models of established micro robots have
been premeditated to accomplish motion Control. The biological and non-biological micro
robots have been Directed through independent planning of motion . As discussed above,
targeting drug delivery mostly relies on The pathological area for the therapeutics’ payload
administration. The systemic route of administration is the most suitable drug Administration
route because most of the drug reaches the Desired area via blood circulation. Due to this
reason, the blood Vessels are made an attractive navigating route to circumnavigate The
wirelessly well-ordered milli/micro robots for the delivery of Targeted theranostic .
Teleoperated microsurgery might include Other therapeutic immersion supplies than
theranostic ones. Like, Micro robots may be employed for thrombolysis and rotational
Atherectomy to clear the inscriptions and blood clotting to maintain blood circulation to the
damaged aorta or other blood vessels. Both benign/malignant tumors acquire their nutrition by
Developing blood vessels (angiogenesis), annexing metastatic cancer cells to limited and
detached areas, and forming a circulatory Network . Marks of malignant tumors (metastasis
and angiogenesis) may be identified explicitly through milli/micro robots and Then block the
antigenic tree at its base to expurgate the vascular tumor’s supply and stop the flow of tumor
cells to distant Organs . Besides, the antigenic system might be employed in A precise body
targeting site to provide milli/micro robots loaded With diagnostics such as (imaging agents),
chemo/radio therapeutic (like drugs, proteins, and radioactive seeds), or genetic agents (mRNA
and DNA) ]. The release of therapeutic materials can be Activated through heating switches or
remote opto-magnetics Milli or micro robots might be employed for scattering radioactive
nucleoids for brachytherapy to destroy the tumors by the Source of radiated energy at the inner
tumor’s core

CHAPTER 4

NANO ROBOTS

It is a type of Nano devices, which are employed for defending or Managing

human pathogens. It is a small instrument synthesized To accomplish a

17
specific project or occasionally tasks with accuracy At nanoscale proportions
of 1–100 nm. They are predictable to Strength at three different atomic,
molecular, and cellular levels to Implement responsibilities in both
therapeutic and industrialize Levels . As stated in Nano robotic theory, ‘nan
robots are microscopic in size, it would probably be necessary for very large
numbers of them to work together to perform microscopic and Macroscopic
tasks’ . Developments in robotics engineering, Computers, medicines,
nanotechnology, and bioinformatics may Lead to the fabrication of smart
nan robotic drug delivery systems. Reciprocyte Nano robots, microbiomes
Nano robots, surgical Nano robots, and cellular repair Nano robots are a few
examples Of Nano robots. These Nano robots are employed for protecting
and defending The human body from disease-causing agents. They have a
width Of about 0.5–3 microns and are fabricated with proportions in 1–100
ranges of nanometers. Carbon, in the form of diamond and Fullerene, is the
most common element utilized in Nano robots Due to its inertness and
strength .. Nano robots have an external passive diamond coating, which is
intended to protect them From assault by the host immune system. Because
they are Undetectable to the human eye, manipulating and working with
Them is difficult. In order to detect the molecular structure of These Nano-
scaled robotic devices, techniques such as Scanning Electron Microscopy
(SEM) and Atomic Force Microscopy (AFM) Are being used. Virtual reality
(VR) procedures are presently being Discovered in biotechnological and
nanoscience to improve the Operator’s awareness by impending extra or less
position of ‘full Immersion’ or ‘telepresence’ . The construction and control
of Nano robots are the two most difficult aspects of developing Nano robots
or Nano machine components. These devices will function In
microenvironments with physical characteristics that are different from
those experienced by traditional components. Nano robotics is a field that
requires a multidisciplinary approach Between biologists, physicists,

18
engineers, computer scientists, Chemists, and other professionals to achieve
this mutual objective. Presently this field is still growing; however, numerous
considerable steps have been taken by experienced researchers and are
Advancing this interesting and exciting field

CHAPTER 4.5
OTHERS

Pharmacyte is a medical nano robot with a diameter of 1–2 lm And a drug

retaining ability up to 1–3 lm in the reservoirs, as Shown in . For
categorisation pumps, they are organised And regulated, employing
mechanical systems. They are delivered By molecular indicators or
chemotactic sensors, which ensure the Precision of the targeting system.
From the limited settings, such As intestinal fluid, blood, and cytosol are the
on-panel supply of Power, glucose, and oxygen extracted. Nanorobots can be
eliminated or recuperated through centrifuge nanapheresis after finishing
the tasks .Microchips Nano robots possess microchips, which are overlapped
by human Molecules, and the chip conducts current signals once the
molecules detect a disease. For example, the distinct sensor nanobots May be
delivered through the blood beneath the skin, validate Blood contents, and
predict infections. They may be applied to Report the level of sugar in the
blood too. The benefits are the Small charge to yield and simple to operate.
Figure 2 shows Nanorobots used for diagnosis and imaging in blood vessels
. Reciprocyte It is a type of Nanorobots that carries oxygen like an artificial
red Blood cell. The power is achieved through endogenous serum glucose.
Such artificial cells can provide 236 times greater oxygen to The tissues per
unit volume than RBCs (Red blood cells) and direct Acidity . Reciprocyte is

19
shown in .Microbivores This device is flat and spheroidal in shape for
nonomedical uses With a diameter of 3.4 lm along its main axis and a
diameter of 2.0 lm on the minor axis . Nano robot may persistently utilized
around 200 PW; this strength is applied to digest stuck Microorganisms.
Another unique character is linked to its phagocytic capability, which is
almost 80-fold higher proficient than Other macrophages regarding
(volume/sec digested) per unit volume of a phagocytic Clottocytes
Clottocytes. Clottocytes have a special biological ability: ‘instant’ hemostasis
Through Clottocytes, named artificial mechanical platelets. Moreover,
platelets are coarsely spherical blood cells without a Nucleus and almost 2
lm in size and stick to a bleeding point and Help in clotting the blood vessels
to stop the bleeding. They also Transport substances that assist in the
coagulation process .Chromallocyte, Chromallocytes would replace whole
chromosomes in individual Cells, repairing the genetic defects and other
chronic damages to The genetic makeup, thus preventing ageing. In a cell,
the renovation machine will first assess the condition by inspecting the
cellular substances, actions, and works along with the structure-by structure
and molecule-to-molecule; these repair machines are Capable of overhauling
the complete cell. Diamondoid shape Nanorobots (cell-repair)

CHAPTER 5

STEPS IN DRUG DELIVERY

Drug delivery using Nanorobots involves several key steps:

1. **Target Identification:**

20
 - Identify the specific site or cells within the body where the drug needs to
be delivered.

- Consider factors like disease location, cellular receptors, or tissue

characteristics.

2. **Nanostructure Design:**

- Design Nanorobots with materials at the nanoscale, often using

biocompatible materials.

- Incorporate drug payloads within the Nanorobots structure.

3. **Functionalization:**

- Functionalize the Nano robot surface with targeting ligands or

molecules that can recognize and bind to the target site.

4. **Navigation:**

- Introduce the nanorobots into the body, allowing them to navigate

through the bloodstream or other bodily fluids.

- Utilize propulsion mechanisms, such as magnetic fields or biological

cues, for targeted movement.

5. **Target Recognition:**

- Exploit the designed recognition elements to specifically identify the

target site or cells.

6. **Drug Release:**

- Trigger the release of the drug payload at the targeted location.

- This release can be controlled by external stimuli (e.g., magnetic fields,

light) or internal factors (e.g., pH, enzymatic activity).

7. **Biodegradation or Elimination:**

21
 - Design nanorobots to be biodegradable or easily eliminated from the
body once their drug delivery mission is completed.

8. **Monitoring and Feedback:**

- Include sensors or imaging components in nanorobots for real-time

monitoring of drug release and therapeutic efficacy.

- Provide feedback mechanisms to adjust drug delivery parameters if

necessary.

9. **Safety and Biocompatibility:**

- Ensure the safety of nanorobots by addressing biocompatibility issues.

- Conduct thorough preclinical studies to assess potential side effects or

toxicity.

10. **Regulatory Approval:**

- Navigate regulatory processes for approval of nanorobot-based drug

delivery systems, ensuring compliance with safety and efficacy standards.

These steps highlight the complexity and precision involved in developing

nanorobots for drug delivery, emphasizing the need for interdisciplinary
collaboration between nanotechnology, medicine, and engineering.

CHAPTER 6

CROP RECOMMENDATION MODELS

Crop recommendation models use a combination of weather, soil, and crop data to provide
farmers with recommendations on the optimal planting, irrigation, and fertilization practices
for a particular crop in a specific region. These models can help farmers maximize their crop

22
yields and minimize their environmental impact by providing tailored recommendations based
on local conditions. There are several types of crop recommendation models, including:

7.1 MACHINE LEARNING MODELS

Machine learning models are a type of artificial intelligence that can learn from large
amounts of data to make predictions and decisions. In crop recommendation, machine learning
models are trained on historical weather, soil, and crop data to identify patterns and
relationships between different variables. These models can then be used to make
recommendations on optimal planting times, irrigation schedules, and fertilizer applications
based on current weather and soil conditions.

Advantages of machine learning models in crop recommendation include their ability to

learn from new data and adapt to changing conditions, their ability to incorporate a wide range
of variables, and their ability to provide personalized recommendations for individual farms.
However, machine learning models can be complex and require large amounts of data to train
effectively.

7.2 DECISION SUPPORT SYSTEMS (DSS)

DSS are computer-based tools that provide farmers with recommendations on the best
management practices for their crops based on local conditions. DSS incorporate weather, soil,
and crop data to make recommendations on planting, irrigation, and fertilization practices. DSS
can also provide information on pest and disease management and help farmers make informed
decisions on the use of pesticides and herbicides.

Advantages of DSS in crop recommendation include their ability to provide real-time

recommendations based on current weather and soil conditions, their user-friendliness, and
their ability to incorporate local knowledge and expertise. However, DSS may be limited by
the availability and accuracy of weather and soil data, as well as the complexity of the models
used.

7.3 EXPERT SYSTEMS

Expert systems are computer-based tools that mimic the decision-making processes of
human experts. In crop recommendation, expert systems use knowledge from agricultural
experts and researchers to provide recommendations on the best management practices for a

23
particular crop in a specific region. Expert systems can also incorporate weather and soil data
to provide personalized recommendations based on local conditions.

Advantages of expert systems in crop recommendation include their ability to incorporate

expert knowledge and experience, their ability to provide personalized recommendations based
on local conditions, and their user-friendliness. However, expert systems may be limited by the
availability and accuracy of expert knowledge and the complexity of the models used.

Overall, crop recommendation models can help farmers make informed decisions on the best
management practices for their crops, thereby maximizing yields and minimizing
environmental impact. The choice of model will depend on the specific needs and resources of
the user, as well as the availability and accuracy of weather, soil, and crop data.

24
 CHAPTER 8

APPLICATIONS OF WEATHER FORECASTING AND

CROP RECOMMENDATION

Weather forecasting and crop recommendation have a wide range of applications, including:

 Agriculture: Weather forecasting and crop recommendation models can help farmers
make informed decisions on the optimal planting, irrigation, and fertilization practices
for their crops. By providing personalized recommendations based on local conditions,
these models can help farmers maximize their crop yields and minimize their
environmental impact.
 Disaster Management: Weather forecasting models can help predict and prepare for
severe weather events such as hurricanes, floods, and droughts. By providing early
warning systems, these models can help reduce the risk of damage to crops,
infrastructure, and human lives.
 Energy: Weather forecasting models can help energy companies optimize the
generation and distribution of renewable energy sources such as wind and solar power.
By predicting weather patterns and energy demand, these models can help energy
companies balance supply and demand and minimize the use of fossil fuels.
 Transportation: Weather forecasting models can help transportation companies plan
routes and schedules based on current and predicted weather conditions. By providing
real-time information on weather conditions, these models can help reduce the risk of
accidents and delays.
 Tourism: Weather forecasting models can help tourists plan their travel and outdoor
activities based on current and predicted weather conditions. By providing accurate and
reliable weather information, these models can help tourists make informed decisions
and avoid unfavorable weather conditions. Overall, weather forecasting and crop
recommendation models have numerous applications in various fields. By providing
accurate and reliable predictions and recommendations, these models can help improve
decision-making, optimize resource utilization, and reduce the impact of weather-
related events on the economy and the environment.

25
 CHAPTER 9
ADVANCEMENT IN TECHNOLOGY
Advancements in technology have revolutionized weather forecasting and crop
recommendation in recent years. Some of the major advancements in technology include:

9.1 REMOTE SENSING

Remote sensing technology uses satellite and aerial imagery to gather data on weather, soil
moisture, and vegetation indices. This data can be used to generate detailed maps of crop health
and growth, which can help farmers make informed decisions on irrigation, fertilizer
applications, and pest management.

9.2 INTERNET OF THINGS (IoT)

IoT technology involves connecting sensors and devices to the internet to collect and
transmit data in real-time. In agriculture, IoT devices can be used to monitor soil moisture,
temperature, and humidity, as well as crop growth and health. This data can be used to generate
real-time recommendations on irrigation and fertilization practices.

9.3 ARTIFICIAL INTELLIGENCE (AI)

AI technology uses machine learning algorithms to analyze large amounts of data and
generate predictions and recommendations. In weather forecasting, AI models can be trained
on historical weather data to predict future weather patterns. In crop recommendation, AI
models can be trained on historical weather, soil, and crop data to generate personalized
recommendations for individual farms.

9.4 BIG DATA

Big data technology involves processing and analyzing large amounts of data to identify
patterns and relationships. In weather forecasting and crop recommendation, big data
technology can be used to identify correlations between weather patterns, soil moisture, and
crop health, which can help generate more accurate predictions and recommendations.

9.5 PRECISION AGRICULTURE

Precision agriculture involves using technology to optimize resource utilization and reduce
waste in agriculture. This includes using precision irrigation systems, variable rate fertilization,

26
and GPS-guided equipment. By optimizing resource utilization, precision agriculture can help
increase crop yields and reduce environmental impact.

Overall, advancements in technology have opened up new possibilities for weather

forecasting and crop recommendation. By leveraging new technologies, weather forecasting
and crop recommendation models can provide more accurate and personalized predictions and
recommendations, helping farmers make informed decisions and maximize their yields while
minimizing their environmental impact.

27
 CHAPTER 10
LITERATURE SURVEY

A. A Web API and Web application development for dissemination of Air Quality
Information (Kevser Sahin, Umid Isikdag)
In this paper the agenda of study is to make a web service for air quality for checking
the quality of air for preventing a dangerous situation. This paper is basically proposed
a system for Air Qualtiy Management in Turkey and also shown the limitations of
previous methods used. Here they have also used the visualization methods for showing
the air quality based on the map of the region. Basically, they have shown how the
basic RESTful Web API development works.
B. Development for REST API for obtaining Site-specific historical and near-future
weather data in EPQ Format (Hu du, Eva Lucas Segarra, Carlos Bandera)
This paper shows how to get real-time current and forecast weather in QPW format
using online toolchain which further can be used as an API for building simulation
community.
C. Java based Webcentric Approach for Weather Prediction Using the ARPS Model
(Gopal Racherla, Adwait Sathye, Anindya Das)
The agenda of this paper is to create a tool to use computers for science rather than to
use computer as a platform and this Java for the ARPS is a step in the same direction
and how this helps in the weather prediction for a particular location.
D. Survey on Weather Forecasting using Data Mining (Christy Kunjumon, Sreelekshmi
Nair, Deepa Rajan, L. Padma Suresh, Preetha S L)
The main purpose of this paper is to show how data mining is used for weather
forecasting and the different types of data mining algorithms like naïve bayes are used
for obtaining more accurate results on the same.
E. Weather Forecasting Using Sliding Window Algorithm (Piyush Kapoor, Sarabjeet
Singh Bedi)
This paper shows how the variations in the weather conditions in the past years be
utilized for predicting the future weather to get more accuracy and efficiency of the
weather forecast.
F. Weather Forecast Prediction: An Integrated Approach for Analyzing and Measuring
Weather Data (Munmun Biswas, Tanni Dhoom, Sayantanu barua)

28
 This paper deals with the weather prediction using chi square and naïve bayes
algorithms where the current weather data is taken into the dataset and extracted to get
useful information for prediction.
G. Weather Forecasting using Machine Learning Algorithms (Nitin Singh, Saurabh
Chaturvedi, Shamim Akhter)
This paper presents a model in which raspberry pie 3 B is used for using ML Algorithms
and then predicting the weather, also some of the hardware like barometer pressure
sensor for calculating the air pressure.
H. Improvement of Crop Production Using recommender system by Weather Forecasts (S
Bangaru Kamatchi, R Parvathi)
The paper proposes a system for best crop recommendation by using weather conditions
and ANN as algorithm showing improvements that can be suitable for the crop
considering the current, historical and future weather.
I. Collaborative Filtering System for Agriculture Sector (Sapna Jaiswal, Tejaswi
Kharade, Nikita Kotambe, Shilpa Shinde)
The agenda of this paper is to use collaborative filtering for giving the best accurate
information to the farmers about the agriculture sector which includes updating farmers
on the recent trends in crop production, government schemes and programs

29
 CHAPTER 11
PROPOSED SYSTEM

We have created a simple web application using JSP, Servlet and MySQL for weather
forecasting and displaying various weather parameters. We have also used API’s for collecting
data and displaying it in the web application for accurate conditions and to make the application
reliable. Basically, three API’s have come across this web application, Open Weather Map,
API, Air pollution and Google Maps API. Weather data in this application includes the current
weather, weather forecast of 5 days, historical weather of 5 days, air pollution data, i.e. the air
quality index. We can also see the weather on different location where we have given a search
option of city in current weather, and latitude and longitudes for weather forecast, historical
weather and air quality index. For crop recommendation, we have used collaborative filtering
which takes the data from the combination of user data and weather data and gives efficient
prediction on the future crop to be grown so that it helps farmers to prevent their land and
produce a healthy crop in his/her land without any damage or any destruction to the land and
the crop.

 Login System
This is a basic login system where the user can create a new account or use
existing account for logging in to see the weather conditions. For creating an
account user has to enter basic information like name, email id, mobile number,
password etc. to save his data into the database.
 Weather components
1) Current Weather This is the welcome page after you log in. This will show the
current weather data according to the device location in which you have
logged in. It has parameters like the city name, temperature, etc. We can also
extract the weather conditions of any city according to our choice using the
search box which takes input as city name and this data is fetched from the API
called Open Weather Map API.
2) Historical weather This is another tab where you can switch from which shows
the historical data, i.e. the weather conditions for the past few days. We can also
see weather conditions in another city using the latitude and longitude of the
particular location and we can track the weather there also.

30
 Weather Forecast This is the main component where we can see the
future weather conditions of the next coming days. Here also the option
of latitude and longitude is given for the user to track weather forecast
of any location.
 Air quality This component takes air as the parameter of weather to
check the air quality. Same as historical weather and weather forecast,
we can use the latitude and longitude.
 Google Maps Images Here, we show google maps and clicking on a
particular point on the map will show a window which will tell display
the user the current weather condition there.
 Crop recommendation This page shows the user which crop to use for
the weather condition according to the location of user’s will. Here
collaborative filtering is used to collect data from all other users and
checking the crops for the suitable weather.

31
 CHAPTER 12
SYSTEM ANALYSIS

This System Analysis is closely linked to the requirements analysis. It's also "an explicit
formal inquiry done to aid someone (referred to as the decision maker) in discovering a better
course of action and arriving at a better conclusion than he would have reached on his own."
Breaking down the system into distinct components to assess the situation, reviewing project
goals, breaking down what needs to be constructed, and engaging users to determine particular
requirements are all part of this process.

I. Functional Requirement Specification:

The System after careful analysis has been identified to be present with the following
modules.
II. Administration Module:
The administrator forms the core of the system. Administrator is responsible for training
the models on the data and loading the model when desired by the user. When a user
submits the input data, the administrator loads the trained model and displays the result.
Another responsibility of the Administrator is to fetch the weather data using the
weather API.
III. User Module:
This module represents the user of our system. The user can submit the input data to
the admin. The data includes the N-P-K, pH values and rainfall. The user then waits for
the response from the Administrator module.
IV. Machine Learning Model Module:
The machine learning model is trained on the dataset. After satisfactory results are
obtained, it is saved for later retrieval. It is invoked by the Administrator and performs
prediction on the data supplied to it.
V. Performance Requirements:
The output supplied by the application is used to assess performance. The specification
of requirements is crucial in the study of a system. It is only possible to develop a
system that will fit into the appropriate environment if the requisite specifications are
properly supplied. Because they are the ones who will use the system in the end, it is
mostly up to the users of the existing system to provide the requirement specifications.
This is because the requirements must be known during the early stages of the project

32
so that the system can be designed to meet them. It is difficult to update a system once
it has been designed, and developing a system that does not meet the needs of the user
is pointless. The current system is entirely reliant on the user to execute all tasks.
For any system can be broadly stated as given below:
o The system must be able to communicate with the existing system.
o The system should be accurate
o The system must outperform the present system.
o The existing system is completely dependent on the user to perform all
the duties.

33
 CHAPTER 13

SYSTEM DESIGN

The process of establishing the architecture, components, modules, interfaces, and data for a
system in order to meet specific criteria is known as systems design. It might be viewed as a
product development application of systems theory. The most extensively used approaches for
computer system design are object-oriented analysis and design methodologies.

a. ARCHITECTURE

Fig 13.1 Block diagram of crop recommendation system

The application is divided into two stages. The first stage is building the required
machine learning model and second stage is deploying it to production. First, we load
the crop dataset, explore the data and perform the necessary feature engineering. After

34
subjecting our data to proper data cleaning, we split it into training and testing sets. We
train the model on the training data and test its performance on our test set.
After choosing the suitable model for our task, we save the data locally to be
accessed by our application when required. The application loads the model that was
saved in the previous phase and waits for user input. The weather data is fetched using
an API. The input data is transformed into a suitable format. The model that was saved
earlier is invoked to get the result. The result is displayed to the user through User
Interface.

35
 CHAPTER 14
SYSTEM IMPLEMENTATION

Any project's implementation stage is a true representation of the defining moments that
determine whether a project succeeds or fails. The system or system modifications are installed
and made operational in a production environment during the implementation stage. After the
system has been tested and accepted by the user, the phase begins. This phase continues until
the system meets the defined user requirements and is ready to go into production.

Sample Code for Our Project:

Data Cleaning/Exploration:

Data Exploration is one of the most important tasks when it comes to tackling a Machine
Learning problem. In order to truly understand the problem at hand, we must understand the
data that we are dealing with. We can be able to view the first and last five rows of our data set
to get an idea of what the features of our data are and how they look.

df= pd. read_csv("Crop_recommendation.csv")

df. head ()

df. tail ()

N P K Temperature Humidity PH Rainfall Label

0 90 42 43 20.879744 82.002744 6.502985 202.935536 Rice
1 85 58 41 21.770462 80.319644 7.038096 226.655537 Rice
2 60 55 44 23.004459 82.320763 7.840207 263.964248 Rice
3 74 35 40 26.491096 80.158363 6.980401 242.864034 Rice
4 78 42 42 20.130175 81.604873 7.628473 262.717340 Rice
Table 14.1 Crop Recommendation Dataset

We must also understand the shape of our data set. That is the number of features and the total
number of records. We use Pandas shape attribute to do this:

df. shape (2200, 8)

This dataset consists of 2200 records and 8 columns. Since this is clearly a classification
problem, it is necessary to know the number of classes in our target label. It is also necessary
to know the total number of records that constitute each class in our data set. We must also

36
check for null values and clean them out if any. To accomplish this, we can use the following
code:

Fig 14.1 Dataset exploration

One of the most popular ways of exploring the relations between the features in our data is
using correlation matrices. Pandas has a built-in correlation matrix for a data frame that we can
use to truly understand the features in our data. By default, the correlation is Pearson’s
Correlation. Here the values range between -1 and +1.

37
 Fig 14.2 Correlation Matrix of Dataset Features

We have assumed a coefficient threshold of 0.8 and hence most of the features are necessary
for us.

Splitting the Data into Training and Testing Datasets:

Before we split our data into training and testing sets, we must separate the features and target
labels that we want to classify the data into. We do this by:

x=df. drop ("label”, axis=1) #Column wise Drop y=df['label']

38
print(x)

print(y)

In order to solve the classification problem, we must split the data into training and testing
datasets. The training data set is the data on which our models will be trained on, and the testing
set is the data in which we evaluate the performance of our model using various metrics. We
use the sklearn module, which has the necessary implementation of how the split is done and
we also import classification_report of sklearn. This will be useful in order to evaluate our
model later on.

from sklearn. model_selection import train_test_split

In the above step, we have split the data such that 20% of the dataset is testing data and the
remaining 80% of the data is training data. As the split is done in a random order, we set the
random_state to 2.

Decision Tree Model:

Decision Trees are a good choice when it comes to solving classification problems. Sklearn
allows us to build Decision Trees in a very fast and time saving manner. We create a Decision
Tree object of the class Decision Tree Classifier. The arguments are as follows: Criterion:

This is how to evaluate the quality of a split. We can choose either ‘gini’ for Gini impurity or
‘entropy’ for information gain. We’ll choose the later for our model. max_depth: This refers to
the depth of our decision tree. We choose a suitable value for our decision tree. random_state:

This refers to the randomness of the features that are permuted at each split.

Decision Tree Classifier (criterion="entropy”, random_state=2, max_depth=5) Now that we

have created an instance of a Decision Tree, we must fit the model on our training data. We
supply the features and the labels of the training dataset as follows. DecisionTree.fit (x_train,
y_train)

Performance of The Model:

Let us now see how our Decision Tree works on test data. For this, we must predict the class
labels for the features of the test dataset. Then we must compare the results obtained with the
actual class labels of the test dataset. Then we can view the performance metrics of the model.

39
A popular metric to do this is the accuracy of the model. To get detailed information, we can
use the classification report provided by sklearn.

predicted_values=DecisionTree.predict(x_test)

X=metrics.accuracy_score (y_test, predicted_values)

print ("DecisionTrees'sAccuracyis: ", X*100)

print (classification_report (y_test, predicted_values))

40
 CHAPTER 15

RESULTS AND DISCUSSION

Fig 15.1 Crop Recommendation System Webpage

The crop recommendation system website was made using HTML, CSS and JavaScript in the
frontend and Flask framework in the backend. The website front end was made using the
Materialize CSS framework. We have used HTML forms to capture the input data from the
user. The ‘Recommend Crop’ button is a submit button that uses the HTTP POST method
which sends the data to the python Flask. Instead of taking humidity and temperature values
from the user as input, we take their city. The Python application uses the OpenWeatherAPI to
get the temperature and humidity values for the corresponding city. The live location button
gets the user’s live location. The live location button invokes the JavaScript code which uses
the geolocation API to get the user’s latitude and longitude coordinates. It then passes these
values to the OpenWeatherAPI to get the name of the city for the give latitude and longitude
values. This is updated in the City field of the forms interface.

41
Fig 15.2 User Data Entered in The Application

42
 CHAPTER 16

CONCLUSION

After looking at the previous works that have been done to forecast weather, including the
study of the papers and planning the project implementation, we can say that the project
provides better accuracy of weather and is more efficient. We can say that weather is one of
the most important factors for human. And how the future weather can make human plan his
day-to-day life. Also, we have used crop recommendation as one of the applications of weather
forecast to show how weather impacts the crops and which crop is suitable for which condition.

The Crop Recommendation System is mainly used to recommend the optimal crop to the
Farmer. With many industries going digital, it is important that the agricultural sector also take
advantage of the various technologies by using them to solve problems faced by the farmers.
Using the approach that we discussed, farmers can expect greater yields. This system can be
integrated with other smart agricultural systems that already exist. It is cost efficient and helps
the farmers make an informed decision.

43
 REFERENCES

1. Kevser Sahin, Umid Isikdag (2017 November) A Web API and Web application
development for dissemination of Air Quality Information. link:
HTTPs://doi.org/10.5194/isprs-annals-IV-4-W4- 373-2017
2. Hu du, Eva Lucas Segarra, Carlos Fernandez (2020). Development of a Rest API for
Obtaining Site-Specific Historical and NearFuture Weather Data in EQW format UK
11-12.
3. G. Racherla, A. Sathye and A. Das, "Java based web centric approach for weather
prediction using the ARPS model," Proceedings of the IEEE 1997 National Aerospace
and Electronics Conference. NAECON 1997, 1997, pp. 271-278 vol.1, doi:
10.1109/NAECON.1997.618090.
4. C. Kunjumon, S. S. Nair, D. Rajan S., P. Suresh and S. L. Preetha, "Survey on Weather
Forecasting Using Data Mining," 2018 Conference on Emerging Devices and Smart
Systems (ICEDSS), 2018, pp. 262-264, doi: 10.1109/ICEDSS.2018.8544326.
5. Kapoor, Piyush & Bedi, Sarabjeet. (2013). Weather Forecasting Using Sliding Window
Algorithm. ISRN Signal Processing. 2013. 1-5. 10.1155/2013/156540.
6. Biswas, Munmun & Dhoom, Tanni & Barua, Sayantanu. (2018). Weather Forecast
Prediction: An Integrated Approach for Analyzing and Measuring Weather Data.
International Journal of Computer Applications. 182. 20-24. 10.5120/ijca2018918265.
7. N. Singh, S. Chaturvedi and S. Akhter, "Weather Forecasting Using Machine Learning
Algorithm," 2019 International Conference on Signal Processing and Communication
(ICSC), 2019, pp. 171-174, doi: 10.1109/ICSC45622.2019.8938211.
8. Kamatchi, S. & Parvathi, R. (2019). Improvement of Crop Production Using
Recommender System by Weather Forecasts. Procedia Computer Science. 165. 724-
732. 10.1016/j.procs.2020.01.023.
9. Sapna Jaiswal, Tejaswi Kharade, Nikita Kotambe, Shilpa Shinde (2019) Improvement
of Crop Production Using recommender system by Weather Forecasts

44