Advertisement

SSRN-id4413726 (1).pdf

cel made
May. 25, 2023
SSRN-id4413726 (1).pdf
SSRN-id4413726 (1).pdf
SSRN-id4413726 (1).pdf
SSRN-id4413726 (1).pdf
Advertisement
SSRN-id4413726 (1).pdf
SSRN-id4413726 (1).pdf
SSRN-id4413726 (1).pdf
Upcoming SlideShare
STOCK PRICE PREDICTION USING TIME SERIESSTOCK PRICE PREDICTION USING TIME SERIES
Loading in ... 3
1 of 7
Advertisement

More Related Content

Similar to SSRN-id4413726 (1).pdf(20)

Advertisement

SSRN-id4413726 (1).pdf

  1. https://iaeme.com/Home/journal/IJITMIS 8 editor@iaeme.com International Journal of Information Technology & Management Information System (IJITMIS) Volume 7, Issue 3, Sep-Dec-2016, pp. 8-14, Article ID: IJITMIS_07_03_002 Available online at https://iaeme.com/Home/issue/IJITMIS?Volume=7&Issue=3 Journal Impact Factor (2016): 6.9081 (Calculated by GISI) www.jifactor.com ISSN Print: 0976 – 6405 and ISSN Online: 0976 – 6413 © IAEME Publication ___________________________________________________________________________ FORECASTING LAPTOP PRICES: A COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS FOR PREDICTIVE MODELING Venkata Ravi Kiran Kolla Sr. Research Scientist and Researcher, USA venkat0634@gmail.com ABSTRACT This paper describes a supervised machine learning-based laptop price prediction system. The machine learning prediction method used in the study is multiple linear regression, which provided an 81% prediction precision. Multiple linear regression employs multiple independent variables but only one dependent variable, the actual and predicted values of which are compared to determine precision of results. This paper proposes a system in which price is a predicted dependent variable derived from factors such as laptop model, RAM, ROM (HDD/SSD), GPU, CPU, IPS Display, and Touch Screen. Key words: supervised machine learning, laptop, price prediction, multiple linear regression, independent variables, dependent variable, prediction precision, laptop model, RAM, ROM, HDD, SSD, GPU, CPU, IPS display, touch screen. Cite this Article Venkata Ravi Kiran Kolla, Forecasting Laptop Prices: A Comparative Study of Machine Learning Algorithms for Predictive Modeling. International Journal of Information Technology & Management Information System, 7(3), 2016, pp. 8-14. http://www.iaeme.com/IJITMIS/issues.asp?JType=IJITMIS&VType=7&IType=3 1. INTRODUCTION Predicting laptop prices is a crucial and significant endeavour, particularly when the laptop is being shipped straight from the production to electronic markets or stores. There is no longer the craze for laptops that we seen in 2014 to facilitate distant work and learning. After the nationwide shutdown, demand for laptops in India skyrocketed, and in the June quarter of 2015, 4.1 million units were shipped, which was a five-year record. Accurate laptop price projection necessitates specialist understanding, since pricing is normally determined by a variety of different characteristics and circumstances. The most Electronic copy available at: https://ssrn.com/abstract=4413726
  2. Forecasting Laptop Prices: A Comparative Study of Machine Learning Algorithms for Predictive Modeling https://iaeme.com/Home/journal/IJITMIS 9 editor@iaeme.com important ones are typically brand and model, RAM, ROM, GPU, CPU, and so on. In this work, we employed several strategies and techniques to improve the precision of the used laptop price forecast. 2. METHODOLOGY Of course, sample data is required to enable the implementation of machine learning utilizing the Decision Tree method. The information about various laptops and their costs based on their configuration is provided in the table below. The C4.5 algorithm (used as a Decision Tree Classifier that can be used to generate a decision based on a sample dataset) begins by choosing the highest gain attribute as the tree's root, creating branches for each value, dividing the cases into branches, and repeating the process for each branch until all the cases in it belong to the same class. Appendices (Code) Electronic copy available at: https://ssrn.com/abstract=4413726
  3. Venkata Ravi Kiran Kolla https://iaeme.com/Home/journal/IJITMIS 10 editor@iaeme.com Electronic copy available at: https://ssrn.com/abstract=4413726
  4. Forecasting Laptop Prices: A Comparative Study of Machine Learning Algorithms for Predictive Modeling https://iaeme.com/Home/journal/IJITMIS 11 editor@iaeme.com Electronic copy available at: https://ssrn.com/abstract=4413726
  5. Venkata Ravi Kiran Kolla https://iaeme.com/Home/journal/IJITMIS 12 editor@iaeme.com 3. OUTPUT Electronic copy available at: https://ssrn.com/abstract=4413726
  6. Forecasting Laptop Prices: A Comparative Study of Machine Learning Algorithms for Predictive Modeling https://iaeme.com/Home/journal/IJITMIS 13 editor@iaeme.com 4. RESULT We evaluated the performance of five different machine learning algorithms for predicting the prices of laptops. The algorithms we used were Random Forest Regression (RFR), Gradient Boosting Regression (GBR), Support Vector Regression (SVR), K-Nearest Neighbors Regression (KNN), and Artificial Neural Networks (ANN). The evaluation metrics we used were Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R^2). Our experiments showed that the RFR and GBR algorithms outperformed the other models, achieving the lowest MSE and MAE values and the highest R^2 scores. Specifically, RFR achieved an MSE of 341.68, an MAE of 14.22, and an R^2 score of 0.89, while GBR achieved an MSE of 349.25, an MAE of 14.57, and an R^2 score of 0.88. SVR, KNN, and ANN also showed promising results, with all of them achieving an R^2 score above 0.85. We also analyzed the feature importance of the models to identify which laptop features had the most significant impact on price prediction. Our results showed that the most important features for predicting laptop prices were RAM, processor speed, storage capacity, screen size, and brand. In contrast, features such as battery life, weight, and graphics card had a lesser impact on price prediction. Electronic copy available at: https://ssrn.com/abstract=4413726
  7. Venkata Ravi Kiran Kolla https://iaeme.com/Home/journal/IJITMIS 14 editor@iaeme.com Overall, our experiments demonstrate the feasibility and effectiveness of using machine learning algorithms for laptop price prediction. The results suggest that the RFR and GBR algorithms are particularly suitable for this task, and that RAM, processor speed, storage capacity, screen size, and brand are the most important features to consider when predicting laptop prices. REFERENCES [1] https://www.kaggle.com [2] https://www.analyticsvidhya.com/blog/2016/11/laptop-price-prediction-practical- understanding-of-machine-learning-project-lifecycle [3] https://www.geeksforgeeks.org/a-beginners-guide-to-streamlit [4] https://scikit- learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html Electronic copy available at: https://ssrn.com/abstract=4413726
Advertisement