Prediction of Computer Type Using Benchmark Scores of Hardware Units (original) (raw)
Related papers
We propose a set of methods to classify vendors based on estimated central processing unit (CPU) performance and predict CPU performance based on hardware components. For vendor classification, we use the highest and lowest estimated performance and frequency of occurrences of each vendor in the dataset to create classification zones. These zones can be used to list vendors who manufacture hardware that satisfy given performance requirements. We use multi-layered neural networks for performance prediction, which accounts for nonlinearity in performance data. Several neural network architectures are analysed in comparison to linear, quadratic, and cubic regression. Experiments show that neural networks can be used to obtain low prediction error and high correlation between predicted and published performance values, while cubic regression can produce higher correlation than neural networks when more data is used for training than testing. The proposed methods can be used to identify suitable hardware replacements.
International Journal of Advance Research, Ideas and Innovations in Technology, 2019
Here in this study, I will analyze the correlation of the buyers with their age and salary using various machine learning classifier algorithms. This study will predict, who will buy a new item faster as soon as it is launched in the market and how it will be related to the age and salary of the people, who are buying it. The aim of this study is to investigate six different types of Machine Learning, Classifier algorithms (namely Logistic Regression, SVM, Naive Bayes, KNN,Decision Tree, Random Forest and to show their comparative analysis and to predict whether a person will buy a certain product as soon as it is launched in the market. Experiments are performed on the Social_Network_Ads data set which is sourced from Kaggle the online community of data scientists and machine learning engineers. The performance of all the above algorithms is evaluated on the various metrics like recall, precision, F1_score and confusion matrix. Results are then compared.
PERFORMANCE ANALYSIS OF SEVERAL MACHINE LEARNING ALGORITHM OR CLASSIFICATION TECHNIQUE A Project
iii ACKNOWLEDGEMENT I am very much grateful to Almighty Allah for the success and final outcome of this project required a lot of guidance and assistance from many people and I am extremely privileged to have got this all along the completion of my project. All that I have done is only due to such supervision and assistance and I would not forget to thank them. I express my deepest respect, gratitude and thanks to my respectable supervisor Mr. Siddikur Rahman, Assistant professor, Department of Statistics, Begum Rokeya University, Rangpur for providing me an opportunity to do the project work and giving us all support and guidance which made me complete the project duly. I am extremely thankful to him for providing such a nice support and guidance, although he had busy schedule managing his professional time. I am very much grateful for his open-door policy for students to consult with him. I owe my deep gratitude to Dr. Md. Roshidul Islam, Chairman, Department of Statistics, Begum
PREDICTING PERFORMANCE OF CLASSIFICATION ALGORITHMS
Classification is the most commonly applied data mining method, and is used to develop models that can classify large amounts of data to predict the best performance. Identifying the best classification algorithm among all available is a challenging task. This paper presents a performance comparative study of the most widely used classification algorithms. Moreover, the performances of these algorithms have been analyzed by using different data sets. Three different datasets from University of California, Irvine (UCI) are compared with different classification techniques. Each technique has been evaluated with respect to accuracy and execution time and performance evaluation has been carried out with selected classification algorithms. The WEKA machine learning tool is used to analysis of these three different data sets based on applying these classification methods to selected datasets and predicting the best performance results.
2023
Customers are the main goal of all business fields, without customers the company will not be able to continue or compete in the business field it is in, even though the company has brilliant products, if it does not have an increase in the number of customers the business will not be able to develop or even go bankrupt. Therefore, it is necessary to make observations and make applications that are able to predict customers who will subscribe so that companies can predict customers who will subscribe correctly without having to wait for confirmation from customers whose possibilities are still unknown. This can be very useful for any company because companies no longer need to look for random customers where it only takes time to find customers. PT. Telekomunikasi Indonesia with its product (Indihome) which is struggling to compete in the business world in the telecommunications and internet sector. Therefore research and development of this application are carried out so that PT. Indonesian telecommunications can get its customers quickly without having to spend a lot of money and effort. Making this application uses a classification method from machine learning technology based on customer historical data. The classification method has many strong algorithms for predicting variables that have more than 1 label. Some of the algorithms used are Logistic Regression, Random Forest Classifier, Support Vector Machine and Decision Tree which are provided by modules in the python programming language, namely SkLearn. The four algorithms will be tested with data balanced using the Oversampling method from the Smote algorithm to get optimal results in automatically predicting customers. Publisher's Note: JPPM stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Performance Comparison and Optimized Algorithm Classification
2020
The current development of technology is quite rapidly not disengaged in a large data processor covering of all areas such as information technology, computer science, medicine, finance and other. This brings a large computing effect in identifying the processing of data. In data analysis for very large data, data processing is very much needed, in this study the authors propose data mining method as a solution to a large data processing problem, data mining is divided into several techniques including classification method techniques that aims to classify large amounts of data to be relevant data information. In this study the authors compared 5 algorithms in the classification method to get better performance in classification problems. Researchers analyze and test 5 Algorithm classifications with 4 different datasets as a tool in the problem of large data classification. .The results of this research show the method SVM is much better to be used 4 comparison methods in calculatin...
International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2023
The "LAPTOP PERFORMANCE PREDICTION" is an important metric used to predict the performance of laptops Laptop manufacturers spend a lot of time, resources, and money designing new systems and newer configurations. Their ability to reduce costs, charge competitive prices, and gain market share depends on how well these systems perform. In this work, concentrate on the system and architectural design processes for parallel computers and develop methods to expedite them. The methodology relies on extracting the performance levels of a small fraction of the machines in the design space and using this information to develop machine learning models to predict the performance of any machine in the whole design space. Laptop performance prediction is useful to accelerate the design space exploration significantly and aid in reducing the corresponding research/development cost and time-to-market.
When computers first came out, they were as big as building rooms and had comparatively modest processing rates. Microprocessor technology led to a significant reduction in computer size and an increase in processing performance. Computers can be made in many different shapes and sizes with different processing powers, depending on their intended uses. In generally the computer systems can be classified on the following basis: Classifications of Computers System:-A. According to Size. B. According to Purposes. C. According to technology. A) According to Size Based on their outward size, internal capabilities, and external purposes, computers are divided into four classes. 1. Supercomputer 2. Mainframe computer 3. Minicomputer 4. Microcomputer SUPER COMPUTER The super computers are the most high performing system. A computer that performs better than a general-purpose computer is called a supercomputer. Supercomputer performance is usually measured in FLOPS (floating-point operations per second) instead of MIPS (million instructions per second). These are specially made to perform multi-specific tasks. Therefore, many CPUs work in parallel order on these supercomputers. This function of a Supercomputer is called Multiprocessing or Parallel Processing. The first supercomputer was created in the 1960s for the American Department of Defence (USA). All of the world's fastest 500 supercomputers run Linux-based operating systems. Supercomputers actually play an important role in the field of computation, and are used for intensive computation tasks in various fields. There are two broad categories of supercomputers: general purpose supercomputers and special purpose supercomputers. General purpose supercomputers can be further divided into three subcategories: 1) Vector processing supercomputers. 2) Tightly connected cluster computers. 3) Commodity computers. Supercomputers designed specifically to accomplish a specific task or objective are referred to as special purpose computers for the opposite reason. Application-Specific Integrated Circuits (ASICs) are usually used by them, and they provide higher performance.
A Comparative Study of Machine Learning Algorithms
2018
The selection of machine learning algorithm used to solve a problem is an important choice. This paper outlines research measuring three performance metrics for eight different algorithms on a prediction task involving undergraduate admissions data. The algorithms that were tested are k-nearest neighbours, decision trees, random forests, gradient tree boosting, logistic regression, naive bayes, support vector machines, and artificial neural networks. These algorithms were compared in terms of accuracy, training time, and execution time.
Evaluation and implementation of machine learning
User satisfaction on web sites depends on many factors and usability is one of these factors. A web site should be organized in a logical manner to aid the user in accomplishing their goal. This paper proposes an automated method for usability testing of a web site using machine-learning techniques to uncover usability issues related to the organization of the pages (information architecture). The information architecture was represented as a weighted directed graph. The graph features are used to train several machine learning models that can then be used to predict the usability score of another website. Models evaluated include classifiers of support vector, random forest, decision tree, regression models, etc. A number of machine learning models are evaluated to determine the best possible model for this specific use case, using 10-fold cross validation on various sized datasets. We also demonstrate a way of extracting a ranked list of prominent features to that can be improved.