Browse Prior Art Database

Chinese term segments based on user input in search engine

IP.com Disclosure Number: IPCOM000240898D
Publication Date: 2015-Mar-11
Document File: 2 page(s) / 89K

Publishing Venue

The IP.com Prior Art Database

Abstract

This invention provides a method to improve the search more accurately while segmenting the words based on user's input. As the Chinese words have ambiguity while segmenting, this solution can help search engine to segment the Chinese words by recording the user input sequence with phrase.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 74% of the total text.

Page 01 of 2

Chinese term segments based on user input in search engine

        他他他他他他他", the experience user can use "他 说的 确实 在理" to make sure the search engine work as expected. But for many other users, if the input string is like above, how to segment the words more accurately?

他他他他他他他"

    "" "" "" "他他" "他他", it can help search engine to understand this sentence more accurately. If the users want to input "他他他他", the most possible sequence maybe "他他" "他他"

Here is the sample for input method while inputting phrase, this diagram show us how user is used to input the phrase he want to search, for sentence

"

               他他他他他他他", the user segments the phrase in his mind and input the phrase "他他" to segment. Oppositely, for sentence "他他他他", the user segments the phrase with "他他"

Diagram 1, the input behavior with input method

By recording this input behavior, the segment system is easy to know how to segment the ambiguity sentence.

Please refer the diagram to understand the process on how this method works.

Currently the search engine uses word dictionary to mapping the Chinese term and segments them to search index. The search result is not quite accurate if the Chinese words have ambiguity while segmenting.

As a experienced user, he uses spaces to segment the words in search engine to improve the search result.

For example "

This invention is to invent a method to improve the search resu...