基于K近邻法的高考录取预测研究

发布时间：2018-04-29 17:37

本文选题：K近邻 + 高考录取　；参考：《湘潭大学》2017年硕士论文

【摘要】：近年来,随着大数据时代的到来,基于数据的统计学习方法已经引起了全世界范围内的广泛关注,通过数据挖掘技术获取的规律和信息已经在包括市场分析,商务管理,工程设计,科学探索等多方面得到了广泛应用。高考一直是国人较为关注的话题,统计学习方法是通过数据挖掘技术,从指定的数据中获取人们所需的相关信息,让人们作出更加实用有效的决策来解决相应实际问题的方法。本文将借助于统计学习方法来对高考进行科学合理的预测。当前,统计学习方法的应用领域十分广泛,诸如语音识别,股票预测分析,生物信息统计等。在统计学习方法中,K近邻方法是一种应用较为普遍的学习方法,并且具有较好的推广。有鉴于此,本文以K近邻算法为主要工具,进行高考成绩的预测分析。在确定好所用方法模型之后,接下来就需要收集数据。为了对高考进行合理且科学的预测,本文以湘潭市某所具有代表性的普通高中2016届高三学生为对象,整理了 652名考生的四次模拟考试成绩和高考成绩,删掉缺考人员的数据后,一共得到520名考生的考试成绩。然后用K近邻算法来分析并处理这些数据,对高考分数和录取批次进行预测。基于收集到的数据集,在实验中,本文主要涉及到了两种情况,分别是高考基础分的预测和高考录取批次的预测。在实验的结尾,将实验结果和2016年高考实际情况进行对比,同时还将K近邻方法下的实验结果和神经网络方法下的实验结果进行了横向对比,结果均显示了 K近邻在高考预测方面具有良好的准确性和可行性。
[Abstract]:In recent years, with the advent of big data's era, data-based statistical learning methods have attracted worldwide attention. The rules and information obtained through data mining technology have included market analysis, business management, etc. Engineering design, scientific exploration and other aspects have been widely used. College entrance examination has always been a topic of concern to Chinese people. The method of statistical learning is to obtain the relevant information from the specified data through data mining technology, so that people can make more practical and effective decisions to solve the corresponding practical problems. This article will use the statistical study method to carry on the scientific and reasonable forecast to the college entrance examination. At present, statistical learning methods are widely used in many fields, such as speech recognition, stock prediction and analysis, bioinformatics and so on. Among the statistical learning methods, the nearest neighbor method is a kind of widely used learning method, and it has a good generalization. In view of this, this paper uses K nearest neighbor algorithm as the main tool to predict and analyze the scores of college entrance examination. After determining the model of the method used, the next step is to collect the data. In order to make a reasonable and scientific prediction of the college entrance examination, this paper, taking a representative senior high school student in Xiangtan city as the object, collates the results of four simulated examinations and the results of the college entrance examination of 652 examinees. After deleting the data of the lack of examiners, a total of 520 examinee scores were obtained. Then the K-nearest neighbor algorithm is used to analyze and process these data to predict the scores of college entrance examination and the batch of admission. Based on the collected data sets, in the experiment, this paper mainly involves two kinds of cases, namely, the prediction of the basic score of the college entrance examination and the prediction of the batch of the entrance examination. At the end of the experiment, the experimental results are compared with the actual situation of the 2016 college entrance examination, and the results of the K-nearest neighbor method and the neural network method are compared horizontally. The results show that K-nearest neighbor has good accuracy and feasibility in the prediction of college entrance examination.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：C81;G639.2

【相似文献】