三、最小二乘法，数据分析的瑞士军刀第二个故事的主角是欧拉(Euler), 拉普拉斯(Lapalace),勒让德Legendre) 和高斯(Gauss),故事发生的时间是十八世纪中到十九世纪初。十七、十八世纪是科学发展的黄金年代，微积分的发展和牛顿万有引力定律的建立，直接的推动了天文学和测地学的迅猛发展。当时的大科学家们都在考虑许多天文学上的问题。几个典型的问题如下：土星和木星是太阳系中的大行星，由于相互吸引对各自的运动轨道产生了影响，许多大数学家，包括欧拉和拉普拉斯都在基于长期积累的天文观测数据计算土星和木星的运行轨道。勒让德承担了一个政府给的重要任务，测量通过巴黎的子午线的长度，海上航行经纬度的定位。主要是通过对恒星和月面上的一些定点的观测来确定经纬度。这些天文学和测地学的问题，无不涉及到数据的多次测量，数据的计算与分析；十七、十八世纪的天文观测，也积累了大量的数据需要进行分析和计算。很多年以前，学者们就已经经验性的认为，对于有误差的测量数据，多次测量取平均是比较好的处理方法，虽然缺乏理论上的论证，也不断的受到一些人的质疑。取平均作为一种异常直观的方式，已经被使用了千百年,在多年积累的数据的处理经验中也得到一定的验证，被认为是一种良好的数据处理方法。以上涉及的问题，我们直接关心的目标量往往无法直接观测，但是一些相关的量是可以观测到的，而通过建立数学模型，最终可以解出我们关心的量。 ...继续阅读 (23)

三、最小二乘法，数据分析的瑞士军刀第二个故事的主角是欧拉(Euler), 拉普拉斯(Lapalace),勒让德Legendre) 和高斯(Gauss),故事发生的时间是十八世纪中到十九世纪初。十七、十八世纪是科学发展的黄金年代，微积分的发展和牛顿万有引力定律的建立，直接的推动了天文学和测地学的迅猛发展。当时的大科学家们都在考虑许多天文学上的问题。几个典型的问题如下：土星和木星是太阳系中的大行星，由于相互吸引对各自的运动轨道产生了影响，许多大数学家，包括欧拉和拉普拉斯都在基于长期积累的天文观测数据计算土星和木星的运行轨道。勒让德承担了一个政府给的重要任务，测量通过巴黎的子午线的长度，海上航行经纬度的定位。主要是通过对恒星和月面上的一些定点的观测来确定经纬度。这些天文学和测地学的问题，无不涉及到数据的多次测量，数据的计算与分析；十七、十八世纪的天文观测，也积累了大量的数据需要进行分析和计算。很多年以前，学者们就已经经验性的认为，对于有误差的测量数据，多次测量取平均是比较好的处理方法，虽然缺乏理论上的论证，也不断的受到一些人的质疑。取平均作为一种异常直观的方式，已经被使用了千百年,在多年积累的数据的处理经验中也得到一定的验证，被认为是一种良好的数据处理方法。以上涉及的问题，我们直接关心的目标量往往无法直接观测，但是一些相关的量是可以观测到的，而通过建立数学模型，最终可以解出我们关心的量。 ...继续阅读 (30)

三、最小二乘法，数据分析的瑞士军刀第二个故事的主角是欧拉(Euler), 拉普拉斯(Lapalace),勒让德Legendre) 和高斯(Gauss),故事发生的时间是十八世纪中到十九世纪初。十七、十八世纪是科学发展的黄金年代，微积分的发展和牛顿万有引力定律的建立，直接的推动了天文学和测地学的迅猛发展。当时的大科学家们都在考虑许多天文学上的问题。几个典型的问题如下：土星和木星是太阳系中的大行星，由于相互吸引对各自的运动轨道产生了影响，许多大数学家，包括欧拉和拉普拉斯都在基于长期积累的天文观测数据计算土星和木星的运行轨道。勒让德承担了一个政府给的重要任务，测量通过巴黎的子午线的长度，海上航行经纬度的定位。主要是通过对恒星和月面上的一些定点的观测来确定经纬度。这些天文学和测地学的问题，无不涉及到数据的多次测量，数据的计算与分析；十七、十八世纪的天文观测，也积累了大量的数据需要进行分析和计算。很多年以前，学者们就已经经验性的认为，对于有误差的测量数据，多次测量取平均是比较好的处理方法，虽然缺乏理论上的论证，也不断的受到一些人的质疑。取平均作为一种异常直观的方式，已经被使用了千百年,在多年积累的数据的处理经验中也得到一定的验证，被认为是一种良好的数据处理方法。以上涉及的问题，我们直接关心的目标量往往无法直接观测，但是一些相关的量是可以观测到的，而通过建立数学模型，最终可以解出我们关心的量。 ...继续阅读 (8)

I have written a new simple blog based Flask, Markdown and Bootstrap, and managed it by github since 2014. You can find it here:www.52nlp.org. Here will be the history of my english blog based on WordPress. ...继续阅读 (18)

It’s time to say goodbye 2013, following is my simple summary for this year.1、MOOC & Open Course:Khan Academy: Keep every day’s practice almostly and following is my record:Code School: Love code school and finished 11 short courses:Code Academy: After 200 days max streak I leave it, because I can’t find any course to work, and finished almost every short course on it:Udacity: I love this MOOC platform and get 5 certificate from it, and now I still learn CS course on it every day.Coursera: I have attended many courses but only get 3 certificate course from it, I think I ...继续阅读 (38)

After 200 Days Streak onCodecademyI will say goodbye to it, cause I have learned everything on it and recently I just repeat to learn I have learned:There are new things I can learn everyday, especially Khan Academy at every morning:Just keep learnig every day! ...继续阅读 (17)

Still in Kehan Academy, I metCentral Limit Theorem, and found that I have wrong thought about CLT, so write a note here for memory. Following definition is from wikipedia:In probability theory, the central limit theorem (CLT) states that, given certain conditions, the mean of a sufficiently large number of independent random variables, each with a well-defined mean and well-defined variance, will be approximately normally distributed. The central limit theorem has a number of variants. In its common form, the random variables must be identically distributed. In variants, convergence of the mea ...继续阅读 (33)

I have made a “1 year plan on Codecademy” and now it’s 100 days streak on Codecademy for me:Posted by52nlp ...继续阅读 (20)

After launchedCourseGraphwhich wrote by Python Flask, I met some nlp problems for English text processing and it’s time to find some Python NLP Open Solutions. And certainly NLTK is the first choice. Following is from Wikipedia:The Natural Language Toolkit, or more commonly NLTK, is a suite of libraries and programs for symbolic and statistical natural language processing (NLP) for the Python programming language. NLTK includes graphical demonstrations and sample data. It is accompanied by a book that explains the underlying concepts behind the language processing tasks supported by the ...继续阅读 (198)

I met the Empirical Rule concept in Kehan Academy, and after google it, I found in statistics, it also known as 68-95-99.7 rule or three-sigma rule, following is fromwikipedia:In statistics, the 68–95–99.7 rule, also known as the three-sigma rule or empirical rule, states that nearly all values lie within 3 standard deviations of the mean in a normal distribution.About 68.27% of the values lie within 1 standard deviation of the mean. Similarly, about 95.45% of the values lie within 2 standard deviations of the mean. Nearly all (99.73%) of the values lie within 3 st ...继续阅读 (17)

I have launched an online education websiteCourseGraphtoday which is used for search, review and recommend open courses and writen by Python Flask. Although it’s not a successul launch cause I met a strange problem in AliYun sever that not exists in the digitocean server. But I think it should be a memorable day in my life, since:Today is Mother’s Day;Today is My BirthDay;Today is One month since I leave Tencent;Today is five years since WenChuan EarthQuake;HopeCourseGraphwill be the best OpenCourse reference for Chinese Students.Posted by52nlp ...继续阅读 (17)

I have learnedCorrelation Coefficientin statistics class when I was a student, but have no impression onCoefficient of Determination or R-Squared. So the first problem for this R-Squared is “What can I do with it?”I find the answerhere:How well does your regression equation truly represent your set of data? One of the ways to determine the answer to this question is to exam the correlation coefficient and the coefficient of determination.Here is two examples how to compute Coefficient of Determination:http://online.stat.psu.edu/online/development/stat501/04linear_assoc/02linear_as ...继续阅读 (17)

I met “Correlation vs. Causality” in Kehan and after google “Correlation an Causality”, I found the Wikipedia give me a reference:Correlation does not imply causation. A lot of refers tell me that Correlation does not imply Causation, following is the defination for them by this website:Correlation vs. CausalityCorrelationis the mathematical relationship between two things which are measured. It is given as a value between 0 and 1. A correlation of 0 means the two things are unrelated; given the first value, there is no way to predict the second. A correlation of 1 m ...继续阅读 (34)

I have learned the “Box-and-Whisker Plots” in Khan Academy, and found it very useful, following is the reference video by Youtube:Following is the text reference I think very good:Box-and-Whisker Plots: Quartiles, Boxes, and WhiskersPosted by52nlp ...继续阅读 (15)