Proceedings of the 2nd International Conference on ICT for Digital, Smart, and Sustainable Development, ICIDSSD 2020, 27-28 February 2020, Jamia Hamdard, New Delhi, India

Research Article

Prediction of Drug Sales by Using Neural Network Algorithm

Download411 downloads
  • @INPROCEEDINGS{10.4108/eai.27-2-2020.2303124,
        author={Ashish  Kumari and Navdeep  Bohra},
        title={Prediction of Drug Sales by Using Neural Network Algorithm},
        proceedings={Proceedings of the 2nd International Conference on ICT for Digital, Smart, and Sustainable Development, ICIDSSD 2020, 27-28 February 2020, Jamia Hamdard, New Delhi, India},
        publisher={EAI},
        proceedings_a={ICIDSSD},
        year={2021},
        month={3},
        keywords={sales prediction random forest xgboost gradient boosting},
        doi={10.4108/eai.27-2-2020.2303124}
    }
    
  • Ashish Kumari
    Navdeep Bohra
    Year: 2021
    Prediction of Drug Sales by Using Neural Network Algorithm
    ICIDSSD
    EAI
    DOI: 10.4108/eai.27-2-2020.2303124
Ashish Kumari1,*, Navdeep Bohra1
  • 1: Maharaja Surajmal Insitiute of Technology,C-4,Janapuri,New Delhi
*Contact email: ashishkumari@msit.in

Abstract

Expectation of sales analysis patterns has been aterritory of extraordinary premium both to scientists end eavoring reveal the data coverd up in the sales information and for the individuals who wish to benefit by predicting sales.The greatly nonlinear nature of the sales information makes it exceptionally hard to structure a framework that can foreseen the future bearing of the sales of articles with adequate exactness. Recommender systems are capable of coming across patterns in sales and generating future income parent based on the patterns as a consequence discovered can considerably supplement the selection-making process of an organization or a trader. Our work presents data analyzation & prediction on sales data by applying neuralnet work algorithms, which produces highly accurate sales forecasts. We have a data that is categorized on various factors i.e. assortment, promotion period, schoolholiday and stateholiday etc.There is also redundant data that is unnecessary as well as some outliers that will be removed too. The data also includes acsv file that contains all the information related to the stores. We feed the data into our trained classifier to give us the prediction of the sales in the coming future, which assist in building a better characteristics and support the sales prediction version.