> For the complete documentation index, see [llms.txt](https://lab-of-ontologies-language-proce.gitbook.io/ladsbook/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://lab-of-ontologies-language-proce.gitbook.io/ladsbook/part-iii-wen-ben-fen-xi-ying-yong-yu-bao-gao/text-clustering.md).

# 文本聚類

* Clustering is the typical **numerical analysis** which tries to group together like observations based on commonality or closeness of the observation data points. 
* In text analysis, we are almost repeating the same operation with clusters: trying to determine the relationships between word usage across a document. **Text clustering** refers to the task of identifying the clustering structure of a corpus of text documents and assigning documents to the identified cluster(s). 
* 常用的方法：Two typical types of clustering algorithms, i.e., connectivity-based clustering (a.k.a., hierarchical clustering) and centroid-based clustering (e.g., k-means clustering).
* `k-means`clustering: reduces the sum of squares differences between relationships and group/cluster words where the distances are minimized to the thresholds specified, in this case, the number of clusters specified.

```r
> library(stats)
> mymeans <- kmeans(dtm,5)
> mymeans
```

```r
> freq <- findFreqTerms(dtm,10)
```

#### 多變量統計