• Nivedita

The efforts are to be estimated before beginning with the development of software. This technique is known as effort estimation. Various models are proposed to measure the efforts accurately. In this work a hybrid formula used that depends upon the values of cost drivers. The generated result will estimate closer efforts than COCOMO. The tool used for the implementation is MATLAB. IVR dataset is being used after taking authentication from the website. This report presented effort estimation models, COCOMO, COCOMO 2, Bailey-Basil model, Halstead, Differential Evolution and the proposed one. In this, there are 63 projects with defined values of cost drivers. The efforts of each model are being calculated and their MMRE is calculated. The result shows that the hybrid model is derived by using TLBO algorithm provides more accurate results than the other estimation models.

Learning Based Optimization (TLBO), COCOMO, Magnitude of Relative Error (MRE).

"Novel Approach to Reduce Magnitude Relative Error Value for Software Effort Estimation", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 8, page no.1089-1094, August-2018, Available :http://www.jetir.org/papers/JETIR1808463.pdf

"Novel Approach to Reduce Magnitude Relative Error Value for Software Effort Estimation", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 8, page no. pp1089-1094, August-2018, Available at : http://www.jetir.org/papers/JETIR1808463.pdf