Obtain relevant property data through the network and preprocess the data, including viewing the correlation between features, removing outliers, performing normal distribution processing, filling in missing values, e...
详细信息
ISBN:
(纸本)9798400709760
Obtain relevant property data through the network and preprocess the data, including viewing the correlation between features, removing outliers, performing normal distribution processing, filling in missing values, etc. Multiple regression algorithms such as elastic regression, lasso regression, ridge regression, and gradient boosting regression were used for modeling, and a stacked model was established on this basis. By comparing the root mean square error of the single model and the stacked models established by different models, the results show that the training effect of the single model is not as good as that of the ensemble model, so the stacked model is selected to predict the house price, which achieves the expected effect.
Glass was introduced to China from West Asia and Egypt through the early Silk Road. Its primary raw material is quartz sand, which has a high melting point. Therefore, during refinement, a flux is added to lower the m...
详细信息
ISBN:
(纸本)9798350375084;9798350375077
Glass was introduced to China from West Asia and Egypt through the early Silk Road. Its primary raw material is quartz sand, which has a high melting point. Therefore, during refinement, a flux is added to lower the melting point. The performance of glass artifacts is influenced by different materials, making the analysis and identification of chemical composition crucial in the study of ancient Chinese glass. In this paper, non-parametric tests and machine learning techniques, including cluster analysis and KNN algorithm, are employed to analyze and identify a collection of ancient glass artifacts. This research investigated the relationship between the weathering condition of glass and its type, patterns, and colors, and predicted the chemical composition of these artifacts prior to weathering. Additionally, a detailed sub-classification and analysis of the chemical composition of glass artifacts is conducted to accurately identify different types of glass, focused on the classification patterns of two specific types of glass, namely high-potassium glass and lead-barium glass as well. Furthermore, the correlation between the chemical compositions of glass artifacts from different categories is also analyzed.
In this paper, body-related data of growth of 1011 healthy yaks as well as the data of their parents were collected from Qinghai Yushu farm. Through analysis it can be found that the yak grows very fast and their body...
详细信息
ISBN:
(纸本)9783030319670;9783030319663
In this paper, body-related data of growth of 1011 healthy yaks as well as the data of their parents were collected from Qinghai Yushu farm. Through analysis it can be found that the yak grows very fast and their body change greatly from one-year-old to three years old, the average of body height, body length, chest girth, circumference of cannon bone and body weight increased percentage can reach 98.46%, 128.1%, 169.39%, 112.34%, and 115.66%. At the age of three, the traits are basically mature, and the yak with excellent traits can be feed as a breeding cattle while the yak with poor traits will be slaughtered. The excellent degree of parental traits has a significant effect on the trait characteristics of descendant yaks. Therefore, parental yaks must undergo strict screening in order to inherit the superior genes to the descendant yaks to a greater extent, which plays an important role in protecting good traits and improving breeding quality. The linearregression model was used to predict the weight of 3-years-old yaks and their parents, the results showed the model fitted well and the accuracy of training set reached 0.84,0.99,0.98 respectively, while the accuracy of test set reached 0.73,0.97 and 0.99 respectively.
This paper aims to predict the weight of yak by analyzing the relationship between body weight and body size characteristics of yak in Three-Rivers Source area of Qinghai Province, and to guide the growth, development...
详细信息
This paper aims to predict the weight of yak by analyzing the relationship between body weight and body size characteristics of yak in Three-Rivers Source area of Qinghai Province, and to guide the growth, development and breeding of yak in this area. In this paper, we selected 303 healthy yaks in Yushu Prefecture of Qinghai Province, the weight and body size(body height, body length, chest circumference, tube circumference) data of yak at different age groups were measured by facilities of weighbridge, measuring stick and tape measure. The data set is divided into training set and test set by random function, cluster the data use k-means clustering algorithm to observe the data distribution. At the same time, Pearson correlation analysis is used to calculate the correlation between body weight and body size, and the training set is analyzed by multiple linearregression and support vector machine prediction. The results show that the average relative error between the predicted value and the true value of the support vector machine is 5.72%, and the average relative error of the multiple linearregression and the true value is 6.52%, which means the support vector machine prediction algorithm can predict the weight of the yak better.
This paper presents a proactive data prefetching mechanism on storage servers for distributed file systems to achieve better input/output (I/O) performance. This mechanism requires keeping tracks of what blocks were a...
详细信息
This paper presents a proactive data prefetching mechanism on storage servers for distributed file systems to achieve better input/output (I/O) performance. This mechanism requires keeping tracks of what blocks were accessed and the information about client I/O requests to predict which blocks will be accessed in the future. Except for showing a linear access tendency, I/O access operations on blocks also reveal a random or similar-random tendency but with certain access patterns. Thus, to address the prediction for future possible read operations, a linearregression algorithm and a chaotic prediction algorithm have been proposed to forecast the future access requests according to different access patterns. As a result, by employing the information, that is, offsets and request sizes, about the predicted future read requests on disk blocks, the storage servers in our newly presented mechanism can prefetch block data more precisely and then send it to associated client file system proactively. The experimental results illustrate that our proposed mechanism is able to shorten read latency to a great extent and then contribute to better read throughput significantly in certain application contexts. Copyright (c) 2014 John Wiley & Sons, Ltd.
暂无评论