tika-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (TIKA-2720) A parser to output universal sentence encodings to text
Date Sun, 02 Sep 2018 22:20:00 GMT

    [ https://issues.apache.org/jira/browse/TIKA-2720?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16601686#comment-16601686
] 

ASF GitHub Bot commented on TIKA-2720:
--------------------------------------

ThejanW opened a new pull request #248: Fix for TIKA-2720 [WIP]
URL: https://github.com/apache/tika/pull/248
 
 
   A parser to output universal sentence encodings to text. This uses Tensorflow Java APIs,
currently have added tests only to verify its abilities. In tests, I mainly shows, how this
parser can be used to output sentence embeddings for multiple sentences all at once. Once
the embeddings are generated, I calculate cosine similarities between each and every sentence
embedding and simply prints out the sentence couples that have high correlations.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


> A parser to output universal sentence encodings to text
> -------------------------------------------------------
>
>                 Key: TIKA-2720
>                 URL: https://issues.apache.org/jira/browse/TIKA-2720
>             Project: Tika
>          Issue Type: New Feature
>          Components: tika-dl
>            Reporter: Thejan Wijesinghe
>            Priority: Major
>             Fix For: 2.0
>
>
> This parser encodes a text into high dimensional vectors that can be used for text classification,
semantic similarity, clustering and other natural language tasks. The model is trained and
optimized for greater-than-word length text, such as sentences, phrases or short paragraphs.
It is trained on a variety of data sources and a variety of tasks with the aim of dynamically
accommodating a wide variety of natural language understanding tasks. The input is variable
length English text and the output is a 512 dimensional vector.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message