A Presentation on

Weather Prediction Using Machine Learning algorithm

------------------------
Presented by
Umesh Yadav(1712010103)
Under the guidance of
Shivangi Dubey(Assistant Professor)
Department of Computer Science and Engineering
Institute of Technology and Management, Gida, Gorakhpur
Date-18/06/2021

Dr. A.P.J. Abdul Kalam Technical University, Uttar Pradesh, Lucknow

 Content

1. Introduction
2. Goals and Motivation
3. Various approaches
4. Literature review
5. Methodology
6. Conclusion
7. References
 Introduction
 The activities of many primary sectors depend on the weather for
production, e.g. farming. The climate is changing at a drastic rate
nowadays, which makes the old weather prediction methods less
effective and more hectic.
 To overcome these difficulties, the improved and reliable weather
prediction methods are required.
 These predictions affect a nation’s economy and the lives of people. To
develop a weather forecasting system that can be used in remote areas is
the main motivation of this work.
 Goals and Motivation
 The main aim of weather forecasting models is to predict temperature,
humidity, dew point, rainfall, and wind speed based on past data.
 Analysis of the literature reveals that several surveys related to wind speed
forecasting have been conducted.
 However, reliable and comprehensive surveys related to forecasting weather
parameters, such as temperature, rainfall, dew point, and humidity, are much
less. Hence, an extensive survey on various weather forecasting models and
their classifications are presented in this paper.
 Climatology and weather forecasting is important since it helps determines
future climate expectations.
 Various Approaches

 Approach 1: Using database for the purpose of handling information as an

integrated whole.

 Approach 2: Using Machine learning-based prediction system, the approach

allows the development of the system to be revised after the stages is
finished.
 Literature Review
Author Name Description
Mohammad Bannayan and "Weather analogue: A tool for real-time prediction of daily weather
Gerrit Hoogenboom, data realizations based on a modified k-nearest neighbor
Environmental Modelling & approach“.
Software, vol. 23, no. 6,
pp. 703-713, 2008.
Kriesche, Bjoern, et al. "A probabilistic approach to the prediction of area weather events,
Spatial Statistics 12 applied to precipitation."
(2015): 15-30.

Chih-Chiang Wei, "Wavelet kernel support vector machines forecasting techniques:

Expert Systems with Case study on waterlevel predictions during typhoons",

Applications, vol. 39,

no. 5, pp. 5189-5199,
2012. [11] Rivero,
Cristian Rodriguez, et
al.
 Methodology

Figure 1: Machine learning-based prediction system.

Several steps are followed for implementing the machine learning
approach in this work. The steps are as follows:
1.1 Data Collection
The purpose of this phase is to determine The data set used for this work is
available at [4]. It is Delhi’s weather data for the past 20 years.

1.2 Data Pre-processing and Feature Extraction

The purpose of this phase is to build All attributes of the data set are processed
once, and the most prominent features of the data set are extracted to be used
further.

1.3 Deriving Another Data set

From the primary data set, another data set was derived in this work by averaging
the temperature, humidity, pressure, and rain for a given date.
1.4 Training and Testing the Machine Learning Model
The random forest classification model is trained with 75% of the derived data
set and 25% is used as the test set for the model. The raw data set had the
details of temperature, humidity, pressure, and rain for each hour of a day. As
shown in Table I, the new data set is derived by averaging the values.

TABLE I. Derived data set from raw data by averaging

1.5 HARDWARE AND SOFTWARE USED
This section deals with the details of the hardware and software used in the work.

Details of Used Hardware:

In this work, the Raspberry Pi 3 B board, DHT11 humidity and temperature sensor, and
BMP180 pressure sensor are used.

Raspberry Pi Board: Raspberry Pi is a computer on a single board. It has advanced in

many aspects, including memory and connectivity of different peripherals. The Raspberry
Pi 3 B is one of the latest versions of Raspberry Pi which includes inbuilt features like
wireless LAN and Bluetooth.

DHT11 Humidity and Temperature Sensor: The DHT11 sensor has the temperature and
humidity sensing capabilities. It provides a digital signal output. The sensor is integrated with an
8-bit microcontroller. This high-performance sensor comprises a resistive element and a sensing
element for negative temperature coefficient measuring devices.
BMP180 Barometric Pressure Sensor: This high precision sensor measures the barometric
pressure or atmospheric pressure. The sensor gives a digital output. It works on inter-
integrated circuit protocol.

Details of Used Software:

The Raspbian operating system (OS), Spyder 3, and Python 3 are used in the system.

Raspbian OS: Raspbian is a computer OS for Raspberry Pi which is Debian based [11]. There
are several versions of Raspbian, including Raspbian Stretch and Raspbian Jessie. It uses
the Pi improved X-window environment, lightweight (PIXEL) as its main desktop
environment.

Spyder 3: Spyder is an open source integrated development environment for Python,

which is significantly used for machine learning [12]. Spyder integrates with many key
packages in the Python stack, such as NumPy, SciPy, Matplotlib, Pandas, IPython, SymPy,
and Cython.
Python 3: Python is a high-level programming language. The automatic memory
management and dynamic type system are its salient features. The object
oriented, functional, imperative, and procedural programming attributes are
supported by Python [13]. It has a significant number of libraries for different
purposes. Python can be easily integrated with many different languages, such as
SQL, MATLAB, C, and Java.
1.6 Proposed work
The tables used in the database are as follow:
1.7 Result
User Interface
 Conclusion
 Weather prediction has drawn a piece of research eagerness in recent years.
The recurrence of customary dangers happening in the light of capricious
climate conditions are unsafe occasions for the human advancement.
 Consequently, in this review different information digging approaches for
climate forecasts were investigated.
 Also, the uses of weather prediction yields positive impact in different fields,
especially agriculture, water system, sun-based power determining and so on
were additionally broke down.
 The examination is made as far as different execution measurements. From
results, it is seen that the Wavelet based SVM and RBF NN utilizing HPSOGA
gives remarkable execution productivity.
 In any case, any enhanced half breed approaches utilizing NN could be
executed for achieving better outcomes.
 References
[1] Mohammad Bannayan and Gerrit Hoogenboom, "Weather analogue: A tool for real-
time prediction of daily weather data realizations based on a modified k-nearest
neighbor approach", Environmental Modelling & Software, vol. 23, no. 6, pp. 703-713,
2008.

[2] Kriesche, Bjoern, et al. "A probabilistic approach to the prediction of area weather
events, applied to precipitation." Spatial Statistics 12 (2015): 15-30.

[3] Matouq, Mohammed, et al. "The climate change implication on Jordan: A case study
using GIS and Artificial Neural Networks for weather forecasting." Journal of Taibah
University for Science 7.2 (2013): 44- 55.

[4] Chih-Chiang Wei, "Wavelet kernel support vector machines forecasting techniques:
Case study on waterlevel predictions during typhoons", Expert Systems with Applications,
vol. 39, no. 5, pp. 5189-5199, 2012.