大侠，帮忙翻译一段文字！！！

Question

在信息技术飞速发展的今天，文本是我们获取信息的主要来源。然而面对海量、异构的文本信息人们往往的显的束手无策。因此，寻找一种行之有效的文本聚类算法已成为一个重要的研究课题。
本文深入研究了几种常用的聚类方法和几种文本数学模型的构建，重点研究了K均值算法及其在文本聚类中的扩展算法—超球K均值算法(SKM算法)。该算法用Salton教授提出的向量空间模型把文本表示成不可视的超球向量空间的一个向量，采用一定的预处理手段(降维、同义词归并、停用词过滤)完成维数约减，使用余弦相似度度量文本之间的相似度，并选择总凝聚度作为目标收敛函数，通过有限次的迭代达到局部最优，输出聚类结果。实验结果验证了SKM算法的有效性和可行性，进一步表明SKM算法简单、高效，十分适合文本挖掘。

Answer 1

In the rapid development of information technology, the text is our today's main source of information. But in front of the text messages, heterogeneous mass of people tend to show the helpless. Therefore, a kind of effective text clustering algorithm has become an important research topic.
This paper studies several common clustering method and mathematical model building several text studied in the text and k-means clustering algorithm of extension algorithm k-means algorithm (super ball SKM algorithm). This algorithm is put forward by professor Salton vector space model of the text representation of a treat of ball of vector space, using a vector of pretreatment methods (dimension reduction, synonyms merging, stop words filter) complete dimension reduction, using cosine about similarity between text and choose the similarity degree of total condensed as the objective function, the convergence of iterative times by limited to local optimal, output clustering results. Experimental results verify the feasibility and effectiveness of the algorithm SKM and further shows SKM algorithm is simple, efficient, very suitable for text mining.

~~~~~~~~~~~~~~~~~~~~~~~希望对你有帮助

Answer 2

Today, the information technology is developed at full speed, the text is the main source that we obtain information. It was apparent and at a loss what to do but often faced magnanimity, different text information people who constructs. So, looking for a kind of effectual text cluster's algorithm already becomes an important subject for research. This text further investigates several kinds of commonly used cluster methods and construction of several kinds of text mathematical models, key research K mean value algorithm and expansion algorithm in the text cluster - exceed the ball K mean value algorithm (SKM algorithm). Algorithm this say text ultra ball vectorial a vector quantity of space that can't look at with Salton vectorial space model that Professor put forward, adopt certain preconditioning means ' lower, link, synonym merge into, stop using word filter), finish dimension reduce nearly, use the similar degree of cosine to measure similar degree between the texts, choose to always condense degree and restrain the function as the goal, reach somely and optimumly through changing taking the place of of limited time, output cluster's result. The experimental result verifies validity and feasibility of SKM algorithm, further indicate SKM algorithm is simple, high-efficient, very suitable for the text to excavate.

Answer 3

Main part being that we gain information stems from today developing at full speed in IT , the version. But version information people who faces great capacity for liquor , the isomerism sometimes obvious no way out. Therefore, seek one kind of effective version cluster-algorithm already becoming a important go into problem. The structure going deep into the mathematic model having studied several kinds clustering method and several kinds versions in common use builds the main body of a book , priority the algorithm having studied K mean value and their expansion algorithm in version clustering surpasses the ball K mean value algorithm (SKM algorithm). A vector owing an algorithm the ball vector space a version is expressed with vector space model brought forward by Salton professor becoming being over being not allowed to look at, adopt certain pretreatment means (conquer the thinking , synonym conflation , stop verbal fliteration) to accomplish a dimension cutting about , uses the similarity degree between degree of amounts cosine similarity versions , chooses general coacervation degree and restraining self's a function as the target, result optimum , exporting clustering by the fact that inferior limited iteration reaches part. The experiment bear fruit has verified algorithmic validity of SKM and feasibility , has indicated the SKM algorithm simplicity further, high-effect, the version very suitable excavates

Answer 4

在信息技术飞速发展的今天，文本是我们获取信息的主要来源。然而面对海量、异构的文本信息人们往往的显的束手无策。因此，寻找一种行之有效的文本聚类算法已成为一个重要的研究课题。
Main part being that we gain information stems from today developing at full speed in IT , the version. But version information people who faces great capacity for liquor , the isomerism sometimes obvious no way out. Therefore, seek one kind of effective version cluster-algorithm already becoming a important go into problem.

本文深入研究了几种常用的聚类方法和几种文本数学模型的构建，重点研究了K均值算法及其在文本聚类中的扩展算法—超球K均值算法(SKM算法)。该算法用Salton教授提出的向量空间模型把文本表示成不可视的超球向量空间的一个向量，采用一定的预处理手段(降维、同义词归并、停用词过滤)完成维数约减，使用余弦相似度度量文本之间的相似度，并选择总凝聚度作为目标收敛函数，通过有限次的迭代达到局部最优，输出聚类结果。实验结果验证了SKM算法的有效性和可行性，进一步表明SKM算法简单、高效，十分适合文本挖掘。
The structure going deep into the mathematic model having studied several kinds clustering method and several kinds versions in common use builds the main body of a book , priority the algorithm having studied K mean value and their expansion algorithm in version clustering surpasses the ball K mean value algorithm (SKM algorithm). A vector owing an algorithm the ball vector space a version is expressed with vector space model brought forward by Salton professor becoming being over being not allowed to look at, adopt certain pretreatment means (conquer the thinking , synonym conflation , stop verbal fliteration) to accomplish a dimension cutting about , uses the similarity degree between degree of amounts cosine similarity versions , chooses general coacervation degree and restraining self's a function as the target, result optimum , exporting clustering by the fact that inferior limited iteration reaches part. The experiment bear fruit has verified algorithmic validity of SKM and feasibility , has indicated the SKM algorithm simplicity further, high-effect, the version very suitable excavates.

这样就可以了。希望能帮到你。

Answer 5

At the present, the text （‘文本’可能有专业词汇，有需要请自行替换）is the main source for us to catch information. However, it is a hard time to deal with such massive and various messages. Therefore, it is necessary and important to find out the most effective way of text clustering algorithm (‘文本聚类算法’为直译，如有需要请替换)。

The paper here has carried out a further research on several clustering methods(‘聚类方法’直译) and mathematical models of text(“文本数学模型”直译) that are widely used nowadays. Specially, it focused its attention upon the K-means algorithm (“K均值算法”为直译)as well as its expanded algorithm in the area of text clustering - the super ball k-means (SKM) algorithm. The SKM algorithm has applied the vector space model（“向量空间模型”为直译)that was made by Professor Salton, which represents the text into an invisible vector of the super ball vector space (“超球向量空间”为直译). Based on the above theory, it had taken four procedures before it was able to output the clustering results. First of all, it needed to complete the process of dimension subtraction through certain pretreatments, such as reducing the dimensions（“降维”是我自行演绎的，如有专属名词请自行替换）, merging the pairs of synonyms（“同义词归并”为自行演绎，如有专属名词请自行替换）, and, leaching the words that has stopped using(“停用词”为自行演绎，如有专属名词请自行替换). After that, it measured the similarity between the texts by the Cosine (“余弦相似度”为自行演绎，如有专属名词请自行替换）. The third step was to select the final condensed degree (“总凝聚度”为自行演绎，如有专属名词请自行替换)of the previous stages as the convergency function of the targets. Finally, it achieved a regional optimum through a limited iteration. And then the clustering results was finally outputted. The results has proved that the SKM algorithm is effective and feasible，which has also been verified as simple, efficient, and, completely suitable for text leaching.