Information processing system, information processing method, and information processing program

• US 10,007,935 B2
• Filed: 02/28/2014
• Issued: 06/26/2018
• Est. Priority Date: 02/28/2014
• Status: Active Grant

• Current Assignee: Rakuten Group, Inc.
• Original Assignee: Rakuten, Inc. (Rakuten Group, Inc.)
• Inventors: Shinzato, Keiji
• Primary Examiner(s): Burke, Jeffrey A
• Assistant Examiner(s): Vu, Thong
• Application Number:
  US14/766,072
• Publication Number:
  US 20160253720A1
• Time in Patent Office: 1,579 Days
• Field of Search: 707727, 707737, 707741, 707767, 707748, 707749, 707728, 707803, 704 1, 704 9, 704231, 704266, 704240
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products

Abstract

The information processing system according to one embodiment includes a specifying unit and an extraction unit. The specifying unit specifies a content word co-occurring with onomatopoeia in one review among a plurality of posted reviews stored in a storage unit. The extraction unit extracts a posted sentence containing the content word from the plurality of posted reviews. In general, posted sentences or posted reviews containing onomatopoeia are likely to include users'"'"' actual experiences. By extracting the posted sentences that contain the content word which is likely to co-occur with onomatopoeia, it is possible to effectively extract the posted sentences on which users'"'"' actual experiences are written.

Claims

1. An information processing system comprising:

at least one memory operable to store program instructions;

at least one processor operable to read said program instructions and operate according to said program instructions, said program instructions including;

acquiring instructions configured to cause at least one of said at least one processor to acquire a list of onomatopoeia by referring to an onomatopoeia database that stores a plurality of onomatopoeia, divide a posted review among a plurality of posted reviews stored in a storage unit into one or more posted sentences, and determine whether each posted sentence contains any onomatopoeia included in the list of onomatopoeia;

specifying instructions configured to cause at least one of said at least one processor to;

obtain a degree of co-occurrence with onomatopoeia, indicating a probability that a given word included in the posted review appears with any onomatopoeia included in the list of onomatopoeia within one posted sentence, by using a ratio of a probability of a presence of a posted sentence that contains both of the given word and at least one onomatopoeia among all of the one or more posted sentences to a product of a probability of a presence of a posted sentence that contains the given word among all of the one or more posted sentences and a probability of a presence of a posted sentence that contains at least one onomatopoeia among all of the one or more posted sentences, and

specify a content word, among the given word, having a certain degree of co-occurrence with onomatopoeia or higher; and

extraction instructions configured to cause at least one of said at least one processor to extract a posted sentence containing the content word from the plurality of posted reviews and output, on a display of a user terminal, via a communication network, the extracted posted sentence containing the content word.

2. The information processing system according to claim 1, wherein

the extraction instructions are further configured to cause at least one of said at least one processor to count the degree of co-occurrence of content words contained in the one posted sentence and thereby obtain a score of the one posted sentence, and then extract the posted sentence with the score of equal to or more than a specified threshold.

3. The information processing system according to claim 2, wherein the extraction instructions are further configured to cause at least one of said at least one processor to output posted sentences with the score of equal to or more than the specified threshold in a descending order of the score.

4. The information processing system according to claim 2, wherein

when a same content word appears a plurality of times in the one posted sentence, the extraction instructions cause at least one of said at least one processor to count the degree of co-occurrence of the content word only once.

5. The information processing system according to claim 2, wherein

each of the content words is a word, and

the extraction instructions are further configured to cause at least one of said at least one processor to count the degree of co-occurrence of the word contained in the one posted sentence.

6. The information processing system according to claim 2, wherein

the content word is a set of a plurality of words, and

the extraction instructions are further configured to cause at least one of said at least one processor to count the degree of co-occurrence of the set of the plurality of words contained in the one posted sentence.

7. The information processing system according to claim 1, wherein the specifying instructions further configured to cause at least one of said at least one processor to process the one or more posted sentences to exclude one or more functional words, and

the degree of co-occurrence with onomatopoeia is obtained by using a ratio of a probability of a presence of a posted sentence that contains both of the given word and at least one onomatopoeia among all of the one or more processed posted sentences to a product of a probability of a presence of a posted sentence that contains the given word among all of the one or more processed posted sentences and a probability of a presence of a posted sentence that contains at least one onomatopoeia among all of the one or more processed posted sentences.

8. The information processing system according to claim 7, wherein the one or more functional words are words that do not carry a lexical meaning, and express grammatical relationships between words.

9. The information processing system according to claim 7, wherein the one or more function words comprises a preposition, a conjunction, an auxiliary verb, an article, and a relative clause.

10. An information processing method performed by an information processing system including at least one processor, the method comprising:

acquiring a list of onomatopoeia by referring to an onomatopoeia database that stores a plurality of onomatopoeia, dividing a posted review among a plurality of posted reviews stored in a storage unit into one or more posted sentences, and determining whether each posted sentence contains any onomatopoeia included in the list of onomatopoeia;

obtaining a degree of co-occurrence with onomatopoeia, indicating a probability that a given word included in the posted review appears with any onomatopoeia included in the list of onomatopoeia within one posted sentence, by using a ratio of a probability of a presence of a posted sentence that contains both of the given word and at least one onomatopoeia among all of the one or more posted sentences to a product of a probability of a presence of a posted sentence that contains the given word among all of the one or more posted sentences and a probability of a presence of a posted sentence that contains at least one onomatopoeia among all of the one or more posted sentences;

specifying a content word, among the given word, having a certain degree of co-occurrence with onomatopoeia or higher; and

extracting a posted sentence containing the content word from the plurality of posted reviews and outputting, on a display of a user terminal, via a communication network, the extracted posted sentence containing the content word.

11. A non-transitory computer readable medium storing an information processing program causing a computer to:

acquire a list of onomatopoeia by referring to an onomatopoeia database that stores a plurality of onomatopoeia, divide a posted review among a plurality of posted reviews stored in a storage unit into one or more posted sentences, and determine whether each posted sentence contains any onomatopoeia included in the list of onomatopoeia;

obtain a degree of co-occurrence with onomatopoeia, indicating a probability that a given word included in the posted review appears with any onomatopoeia included in the list of onomatopoeia within one posted sentence, by using a ratio of a probability of a presence of a posted sentence that contains both of the given word and at least one onomatopoeia among all of the one or more posted sentences to a product of a probability of a presence of a posted sentence that contains the given word among all of the one or more posted sentences and a probability of a presence of a posted sentence that contains at least one onomatopoeia among all of the one or more posted sentences;

specify a content word, among the given word, having a certain degree of co-occurrence with onomatopoeia or higher; and

extract a posted sentence containing the content word from the plurality of posted reviews and output, on a display of a user terminal, via a communication network, the extracted posted sentence containing the content word.