This report is the final year report of Mr. Prabi Thapa. He is a graduate of university of Wolverhampton. We will provide you the link to his project very soon. You can download his report by clicking download here button at the end of this page.
Calculation of mathematical equations has been done using various method or technology. Abacus was the device used at the very beginning. It consists of series of rod having ten sliding beads on each. Other beads were only moved when all the beads of first rod is moved. Addition and subtraction was much faster. Later on slide ruler is developed to perform calculations. It uses the logarithmic scale to perform multiplications and division.
Later on the electronic calculator was developed. It was developed in late 1930s due to the war. Trigonometry was calculated to drop bombs from 30,000 feet ‘into a pickle barrel’, to hit a Japanese warships with a torpedo. Then the calculator is developed named ENIAC (Electronic Numerical Integrator and Computer) that could solve four basic athematic functions. The world’s first all-electronic calculator was ANITA (A New Inspiration to Arithmetic/Accounting). Later the digital calculator was developed that could do different operations. (Valentine, 2019)
Nowadays, mobile devices and digital calculator being used to perform the small and large calculations. We can use our mobile device to do the calculation. With the help of CNN (Convolution Neural Network) we can develop the mathematical equation solver. We can develop the model by feeding the datasets to the neural network and train the model to obtain better accuracy.
Mathematics is practices of measuring, counting and defining the shapes of objects that is evolved to science of structure, order and relation. Mathematics deals with quantitative calculation and logical reasoning. It has become necessary supplement since the 17th century also assumed as vital role in quantitative aspects of life science. (Gray, et al., 2020)
Mathematics is also concerned to the intuition, imagination and reasoning to solve puzzle and finding new ideas. (Khan, 2015) Including the relationship, measurements and operations mathematics is defined as the scientific study of quantity expressed by numerical values and symbols. (Yadav, 2017)
Math Equation solver helps to recognize the digits and symbols that are handwritten. It calculates the equation after it is getting recognized. It provides the solution after performing certain operations in the system. This system helps student to know the correct answer of their solution and help them identify the mistake. To recognize the digit the result is processed in vote by CNN. With the help of CNN the system can recognize the digits and operators as in CNN the error gets calculated each time and only those adjustment were kept that makes less error. Therefore it helps to get the correct digit and operators. And it can be used to solve simple mathematical equation by scanning them. (Brandon, 2016)
- Sometimes students solves the problem but cannot know the correct answer.
- Some parents may be less educated and they might not know to guide their children. iii. Student may feel disappointment if they did not know the right answer.
As nowadays many children or students uses mobile devices for entertainment except that these devices can be used to for learning purpose as well. I have chosen the project mathematical equation solver so that students would get facility to check the answer of their equations without depending on others. If they can check that the solution is correct or not they can move to solve other equations with confident. If the students are not sure about the correct solution then they might get demotivated and leave the work.
- How the digit and operators recognition will be done?
- How the AI will be deployed in Android?
- What methodology will be used to develop the system?
The aim of the project is to create the system to solve the simple mathematical equation. It intends to make student learn themselves even if there is no guidance and check their answers. The system is about to recognize the digits and symbol and provide the final answer.
i. Create login dashboard
ii. Collect the required dataset
iii. Research on algorithms and the similar system that are developed iv. To create a system that recognize handwritten digits and operators. v. To calculate the problem and display the final solution in screen.
i. to clearly define about computer vision and how it works
ii. create better accuracy model
iii. to provide services to user
iv. Develop the UML diagram and flow chart on the basis of project.
v. Understand the working flow of CNN.
vi. Test the performance of the system
vii. to complete report on time
Mathematical equation solver is an android application which will be developed using java and python where AI model will be created using python on Jupyter notebook and app development will be done on Android Studio.
The system is developed to recognize the digits and operators that are written by hand. The dataset from the Kaggle will be used. 0-9 and +, – and * symbol will be downloaded from Kaggle. Those dataset will be trained to create a model that can be used in app.
Mathematical equation solver solves the simple equations that are in hand written form. User need to scan over the handwritten equations and the app recognize the digit and operators and calculates the solution and return final result on the screen.
The project is developed with the certain scope that it can handle and work with. It can work with the limited numbers of dataset for example 0-9 and +, – and * operators. This system is developed to perform certain activities remaining in its scope.
- To recognize the digit from 0 to 9 and the symbol like +, – and *.
- Scanning the image containing simple equation and the system will solve the mathematical equations.
- Project is not trained to recognize alphabet, brackets, / and other symbol than mentioned above.
- The final solution is displayed on the screen rather than the solution steps.
- Cannot solve the equation that is in words.
- Solves only simple equations.
This section contains the introduction part of the report. Overall introduction of the system, history or the evolution of technology, problem statement, aims and objectives and artifact to be developed are included in this section.
- Literature Review
In this section the review of similar system is done and discussed. Further the problem domain is discussed along with their solution. Here the background research on the similar system will be done and feature will be included. The AI algorithm and its component to be used in the system is discussed. Technical term about the AI will be discussed in this section.
The SDLC, tools and technique are considered and used in this section. How the system will be developed with the methodology is mentioned here in the section. Flowchart, use case diagram, ER Diagram and prototype is developed.
- Implementation (Digit recognition)
The digit recognition is done in this section. With the trained model the prediction is done to recognize the digits and operators.
- Implementation (Android Application)
The work flow of android application is shown in this section. The wireframe and the layout of the system is presented here.
Testing the android application and the digit recognize system is done in this section. The aim of this section is to make sure that the system works fine by finding out the bugs and mistake in the system. Also the requirement to be fulfilled will be identified.
The overall conclusion of the project is done by summarizing about the system developed and possibilities in future of the project is discussed in this section. What changes can be made to improve the system is mentioned in this section.