Answer the question
In order to leave comments, you need to log in
Search for keywords within a sentence. Has anyone gone into detail?
It would be very interesting to hear people and get qualified
comments on the following issue:
There is a large array of incoming text (Russian and English),
divided into sentences. After processing the text at the output, it is necessary
to obtain the highest quality tags (keywords)
for this sentence from the point of view of a person.
As an example,
Input: “My uncle has the most honest rules, when he seriously fell ill ...”
Output: [uncle], [sick]
The topic probably has an endless area for development and discussion,
so the following points are of particular interest:
- the most successful / promising theoretical approaches and directions
— effective "open" tools/algorithms available in personal research.
Answer the question
In order to leave comments, you need to log in
This question worries many. Start with a wiki (http://ru.wikipedia.org/wiki/TF-IDF), polish with google until done.
Judging by the example given by the author of the question, it is enough just to highlight the SUBJECT and the PREDICT in the original sentence.
Yandex has a ready-made free tool MYSTEM.EXE for morphological analysis of sentences in Russian.
[http://company.yandex.ru/technology/mystem]
if run without parameters for this example, it will give:
my{my|wash}uncle{uncle}of the most{most}honest{honest|honest}rules{rule| edit}when
As an option, experiment by counting the most frequently used words in Yandex statistics wordstat.yandex.ru/?. There must be several factors ...
I don’t know the purpose of your undertaking, but for some reason it seems to me that the keywords for each sentence are too much. Why not in paragraphs?
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question