逻辑斯蒂回归模型在电信领域中的应用
|
摘要
为解决电信领域中新服务的接受率问题,引入逻辑斯蒂回归模型。通过分析电信领域注册用户的类别特点,选取影响接受率的关键因素作为因变量,导出罗杰斯谛回归模型。在线性回归模型的基础上,使用sigmoid函数,将线性模型w Tx的结果压缩到[0,1]之间,使其拥有概率意义,最后利用牛顿法迭代出最大似然估计,求解出最优解。实验证明,逻辑斯蒂模型在实际应用过程中具有较高的识别率,有令人满意的识别速度及鲁棒性。
In order to solve the acceptance rate of newservice in the field of telecommunications,the logistic regression model is introduced. By analyzing the category characteristics of registered users in the telecom field,the key factors that affect the acceptance rate are selected as the dependent variables,and the Rogers regression model is derived. On the basis of the linear regression model,the sigmoid function is used to compress the results of the linear model wtx to [0,1 ],so that it has the probability significance. Finally,the maximum likelihood estimation is iterated by Newton method,and the optimal solution is obtained. The experimental results showthat the Rogers model has a high recognition rate in the process of practical application,and has satisfactory recognition speed and robustness.
In order to solve the acceptance rate of newservice in the field of telecommunications,the logistic regression model is introduced. By analyzing the category characteristics of registered users in the telecom field,the key factors that affect the acceptance rate are selected as the dependent variables,and the Rogers regression model is derived. On the basis of the linear regression model,the sigmoid function is used to compress the results of the linear model wtx to [0,1 ],so that it has the probability significance. Finally,the maximum likelihood estimation is iterated by Newton method,and the optimal solution is obtained. The experimental results showthat the Rogers model has a high recognition rate in the process of practical application,and has satisfactory recognition speed and robustness.
引文
[1]陈倬.基于逻辑回归的多任务域快速分类学习算法[J].数字技术与应用,2016(11):123-123.
[2]顾鑫,曹丹华,吴裕斌,等.基于逻辑回归的多任务域快速分类学习算法[J].计算机工程与应用,2017(15):47-56.
[3]张鹏丽,李育.逻辑斯蒂模型在河谷型城市洪水事件研究中的验证[J].西北师范大学学报:自然科学版,2017,53(1):128-134.
[4]杨波,余建星,王谦.逻辑斯蒂增长模型在集装箱运量长期预测中的应用[J].海洋技术,2006,25(4):88-93.
[5]梁慧玲,王文辉,郭福涛,等.比较逻辑斯蒂与地理加权逻辑斯蒂回归模型在福建林火发生的适用性[J].生态学报,2017,37(12):4128-4141.
[2]顾鑫,曹丹华,吴裕斌,等.基于逻辑回归的多任务域快速分类学习算法[J].计算机工程与应用,2017(15):47-56.
[3]张鹏丽,李育.逻辑斯蒂模型在河谷型城市洪水事件研究中的验证[J].西北师范大学学报:自然科学版,2017,53(1):128-134.
[4]杨波,余建星,王谦.逻辑斯蒂增长模型在集装箱运量长期预测中的应用[J].海洋技术,2006,25(4):88-93.
[5]梁慧玲,王文辉,郭福涛,等.比较逻辑斯蒂与地理加权逻辑斯蒂回归模型在福建林火发生的适用性[J].生态学报,2017,37(12):4128-4141.