Automated Software Cost and Estimation System

In the proposed system, an ensemble model is proposed to improve the estimation accuracy of software development effort prediction. It integrates approaches such as Use Case Points (UCP), expert judgement, and Case-Based Reasoning (CBR).  In addition, we have examined a systematic review

Project Title

Automated Software Cost and Estimation System

Project Area of Specialization

Software Engineering

Project Summary

Project Summary:

In the proposed system, an ensemble model is proposed to improve the estimation accuracy of software development effort prediction. It integrates approaches such as Use Case Points (UCP), expert judgement, and Case-Based Reasoning (CBR). 

In addition, we have examined a systematic review of studies on use case points and expert judgment-based software development effort estimation. Furthermore, a comparative study is conducted concerning effort estimation accuracy in solo and ensemble techniques. The estimation accuracy prediction of the proposed model is evaluated by using projects from software development organizations and student projects as case studies in a quantitative manner that includes experiments, industrial experts, archival data on estimates, and evaluation criteria. Software development firms and practitioners will use the proposed system as an instrument to estimate their software’ effort

Project Objectives

Objective:

The proposed system aims to implement an ensemble model to improve the estimation accuracy prediction of software development effort. The ensemble model is incorporated with Use Case Point (algorithmic), expert judgement (non-algorithmic) and Case-Based Reasoning (machine learning) techniques to make an ensemble. Hence, the objectives are as follows to achieve this aim:

1. O1: To investigate the effort estimation models and the accuracy improvement of ensemble and solo techniques.
2. O2: To provide automated solutions to accurately estimate software project effort.
3. O3: To evaluate the improvement of estimation accuracy prediction of an ensemble model by comparing existing solo models.
4. O4: To validate the applicability of the proposed system with diverse evaluation metrics in software development

Project Implementation Method

Platform:

The targeted platform for the system will be the web. The reasons for choosing the we as the platform are following.

R1: Ease of accessibility, web is OS independent anyone with a modern browser can easily and fully access the system.

R2: Speed, Web apps are significantly lighter than native install-able software. The is no installation required to get started, hence no disk space is consumed. Additionally, all the CPU intensive calculations are performed by the server keeping the client’s machine free from any of this processing.

R3: Development speed, When compared with native software development, web applications can be created much faster. Even if using a platform independent language like Java, there is still much that has to be configured for each different operating system. There are library version conflicts that make the development process slower. Developing a web app means coverage for all operating systems (that can run a modern browser of-course) for example Linux, Mac-OS, Windows, Open-BSD or any other UNIX based operating system

Technologies:

We have chosen MERN (MongoDB, ExpressJS, ReactJS, NodeJS) as our stack of technologies to develop the proposed system. MERN is no doubt one of the most popular web development stacks currently. People are shifting from older technologies like Laravel and Django to ExpressJS for creating web services, just because of the ease of use and intuitive syntax of the JavaScript language.  Following is a list of reasons for choosing MERN stack.

R1: Non-relational or No-SQL database. MongoDB is what’s called a “Non-Relational” database. Unlike SQL database, it doesn’t have rows, columns or even tables. Rather data is stored in JSON format like objects. This approach of storing data comes with various benefits like flexible database schema, ability to store unstructured, semi-structured, or structured data, and handle large volumes of data at high speed with a scale-out architecture.

R2: Single Page Application. SPAs or Single Page Applications are websites that act like native applications. They do not require reloading for any operation, so they experience is just like using a mobile applications or native desktop application. Aside from that, their performance is considerably faster than simple HTML or PHP generated pages, this is because of the “virtual DOM” that is rendered using JavaScript. By utilizing the virtual DOM we can choose to only update the parts of the page that have received an update in their state. This prevents unnecessary renders hence improving performance

R3: JavaScript back to front. Every one of the four technologies in the MERN stack is based on JavaScript. This makes a great ecosystem to work in. Instead of developing and maintaining code base in 2 or even 3 languages, we only have to worry about JavaScript. This way we can focus more on the application logic instead of worrying about learning syntax for different Programming Language

Benefits of the Project

We are solving a problem that the entire software development industry is facing. Every software development firm would like an estimate of how much this project will cost them, from there they can calculate their profits and give the client an accurate estimate. Such an estimate that they do not have to fear about going over budget.

Another source of motivation is that there is no competition at all in the industry for such a product. We are the pioneers of automating Software Cost Estimation and creating a commercial product for it.

This is the key motivation and benefits of our peoject which are able to implements in the field of Software Engineering. From our Proposed Solution, the estimation of cost and effort of the upcoming projects will be nearby calculated.

Technical Details of Final Deliverable

Our system is based on multiple technologies to help divide the whole system into smaller manageable sub systems. Following is a list of all logical sub subsystems of the proposed system.

Front-end:

This is the most viewed, hence the most important part of the system. This is where all the interaction of the users will happen. We have decided to develop this part of the system into an SPA or Single Page Application instead of a traditional web page. The reasons for following this approach are mentioned below.

Back-end:

    This is where all the processing will take place. This is where our custom machine learning model will be implemented. This layer will be connected to a database as well as a data warehouse.

RESTful API:

A REST API (also known as RESTful API) is an application programming interface (API or web API) that conforms to the constraints of REST architectural style and allows for interaction with RESTful web services. This layer will be responsible for transporting our data from the server to the front-end layer.

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Finance

Core Technology

Big Data

Other Technologies

Artificial Intelligence(AI)

Sustainable Development Goals

Partnerships to achieve the Goal

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com