improving Readme file

README.md

@@ -1,6 +1,29 @@
 # Singularity-POS-Tagger
 Portuguese POS-Tagger writen in core Node.JS, without any external modules. 
 
+I developed this library to use as base for another personal project. There are planing of room to improve accuracy with heuristics and twiks.
+
+It's designed specially for Node.Js Streams, which can improve speed and memory use when working on servers or large corpus of data. Nonethless, one can still use a built in method to work with strings.
+
+There is no need to pre-processing the corpus, there is a built in function that cleans everything before the POS classification.
+
+Because the nature of JavaScript be single threaded and NLP jobs are usually very resource intense, almost everything in this package runs asyncronous.
+
+# Installation 
+In a Node.JS environment you can run on your terminal. 
+>npm i singularity-tagger
+
+# What is a POS Tagger?
+
+A POS Tagger or Part of Speach Tagger is a piece of software that analyzes a corpus and taggs the words with it's respective gramatical class. 
+
+# Applications of POS Taggers
+  - Sentiment analysis
+  - Question answering
+  - Word sense disambiguation
+
+Basically every Natural Language Processing task uses a POS tagger as sub task.
+
 # Implementation 
 - Model trained on Mac-Morpho's anotated corpus available on: http://nilc.icmc.usp.br/macmorpho/
 - Stochastic algorithm used: 
@@ -13,7 +36,7 @@ Portuguese POS-Tagger writen in core Node.JS, without any external modules.
 *__Singularity__* is designed to be used on async functions or as ECMS6 promise.
 There are two main methods available:
 - __analyzeString__
-    * receive as parameter one string and returns an array with the words normalized 
+    * receive as parameter one string and returns an array with normalized words
     along side the tags of the string
 - __analyzeStream__
     * receive as parameter an input stream and output stream and returns to the output stream the normalized text along side the tags
@@ -49,4 +72,8 @@ There are two main methods available:
         >
         >...
         >
-        > PosTagger().then((tagger)=> tagger.analyzeStream(inputStream, outputStream)).catch(err=>console.log(err))
+        > PosTagger().then((tagger)=> tagger.analyzeStream(inputStream, outputStream)).catch(err=>console.log(err))
+  
+# Tags meaning table
+
+![image info](./assets/table.png)