Server for automatically scoring opinion conveyed by text message containing pictorial-symbols

• US 8,478,582 B2
• Filed: 02/02/2010
• Issued: 07/02/2013
• Est. Priority Date: 02/04/2009
• Status: Active Grant

• Current Assignee: KDDI Corporation
• Original Assignee: KDDI Corporation
• Inventors: Habu, Yukiko, Kawada, Ryoichi, Kotsuka, Nobuhide, Jiae, Sung, Uchiyama, Koki, Saeyor, Santi, Asano, Hirosuke, Shimamura, Toshiaki
• Primary Examiner(s): GUERRA-ERAZO, EDGAR X
• Application Number:
  US12/698,486
• Publication Number:
  US 20100198584A1
• Time in Patent Office: 1,246 Days
• Field of Search: 704/9, 704/10, 704/256, 704/258, 704/260, 704/270, 704/270.1, 704/275
• US Class Current: 704/9
• CPC Class Codes: G06F 40/237 Lexical tools

Abstract

A server is disclosed for computing a score of an opinion that a message in a text file is expected to convey regarding a subject to be evaluated, wherein the message is written using literal strings and pictorial symbols. In this server, by the use of a pictorial-symbol dictionary memory storing a correspondence between designated pictorial-symbols to be rated and scores of opinions expressed by the respective pictorial-symbols, at least one of the used pictorial-symbols in the message which is coincident with at least one of the designated pictorial-symbols stored in the pictorial-symbol dictionary memory, is extracted from the message, at least one of the opinion scores which corresponds to the at least one extracted pictorial-symbol is retrieved within the pictorial-symbol dictionary memory, and an aggregate net opinion score for the message is calculated, based on an aggregate opinion score for the at least one extracted pictorial-symbol.

Claims

1. A server for computing an opinion score which is a score of an opinion that a message in a text file is expected to convey regarding a subject to be evaluated, wherein the message is written using literal strings and pictorial symbols, the server comprising:

a pictorial-symbol dictionary memory storing a correspondence between designated pictorial-symbols to be rated and scores of opinions expressed by the respective pictorial-symbols;

a pictorial-symbol extractor configured to extract from the message, at least one of the used pictorial-symbols in the message which is coincident with at least one of the designated pictorial-symbols stored in the pictorial-symbol dictionary memory;

a pictorial-symbol opinion-score retriever configured to retrieve within the pictorial-symbol dictionary memory, at least one of the opinion scores which corresponds to the at least one extracted pictorial-symbol; and

an opinion-score calculator configured to calculate an aggregate net opinion score for the message, based on an aggregate opinion score for the at least one extracted pictorial-symbol, wherein the aggregate opinion score is calculated with the at least one retrieved opinion score for the at least one extracted pictorial-symbol,

wherein the pictorial-symbol dictionary memory stores an additional correspondence between at least one selected pictorial-symbol and at, least one intensity rate when the at least one selected pictorial-symbol is used as a linguistic modifier for describing another pictorial symbol on which the at least one selected pictorial-symbol depends linguistically, and

the pictorial-symbol extractor extracts from the message, as a modificative pictorial-symbol, at least one of the used pictorial-symbols in the message which is coincident with at least one of the at least one selected pictorial-symbol stored in the pictorial-symbol dictionary memory,

the server further comprising;

an intensity-rate retriever configured to retrieve within the pictorial-symbol dictionary memory, at least one of the at least one intensity rate which corresponds to the at least one extracted modificative-pictorial-symbol; and

a linguistic-dependency analyzer configured to detect from the message, as a depended pictorial-symbol, at least one of the used pictorial-symbols in the message on which each of the at least one extracted modificative-pictorial-symbol depends linguistically,

wherein the opinion-score calculator calculates the aggregate net opinion score for the message, by multiplying the opinion score for each of the at least one detected depended-pictorial-symbol, by the intensity rate for each of the at least one extracted modificative-pictorial-symbol, wherein the opinion score and the intensity rate are defined according to the correspondence stored in the pictorial-symbol dictionary memory.

2. The server according to claim 1, further comprising a morpheme analyzer configured to segment the message into morphemes, wherein the pictorial-symbol extractor extracts at least one of the morphemes of the segmented message, which corresponds to at least one of the designated pictorial-symbols stored in the pictorial-symbol dictionary memory.

3. The server according to claim 1, further comprising:

a literal-string dictionary memory storing a correspondence between designated literal-strings to be rated and scores of opinions expressed by the respective literal-strings;

a literal-string extractor configured to extract from the message, at least one of the used literal-strings in the message which is coincident with at least one of the designated literal-strings stored in the literal-string dictionary memory; and

a literal-string opinion-score retriever configured to retrieve within the literal-string dictionary memory, at least one of the opinion scores which corresponds to the at least one extracted literal-sting,

wherein the opinion-score calculator calculates the aggregate net opinion score for the message, based on a sum of a first aggregate opinion score for the at least one extracted pictorial-symbol, and a second aggregate opinion score for the at least one extracted literal-string, wherein the first aggregate opinion score is calculated with the at least one retrieved opinion, score for pictorial symbol, and the second aggregate opinion, score is calculated with the at least one retrieved opinion score for literal string.

4. The server according to claim 1, wherein each of the opinion scores is varied between discrete possible values, corresponding to a positive opinion and a negative opinion.

5. The server according to claim 1, wherein the server is a Web server for use in mobile phones.

6. A method of computing an opinion score which is a score of an opinion that a message in a text file is expected to convey regarding a subject to be evaluated, wherein the message is written using literal strings and pictorial symbols, the method comprising the steps of:

by the use of a pictorial-symbol dictionary memory storing a correspondence between designated pictorial-symbols to be rated and scores of opinions expressed by the respective pictorial-symbols, extracting from the message, at least one of the used pictorial-symbols in the message which is coincident with at least one of the designated pictorial-symbols stored in the pictorial-symbol dictionary memory;

retrieving within the pictorial-symbol dictionary memory, at least one of the opinion scores which corresponds to the at least one extracted pictorial-symbol; and

calculating an aggregate net opinion score for the message, based on an aggregate opinion score for the at least one extracted pictorial-symbol, wherein the aggregate opinion score is calculated with the at least one retrieved opinion score for the at least one extracted pictorial-symbol,

wherein the pictorial-symbol dictionary memory stores an additional correspondence between at least one selected pictorial-symbol and at least one intensity rate when the at least one selected pictorial-symbol is used as a linguistic modifier for describing another pictorial symbol on which the at least one selected pictorial-symbol depends linguistically, and

the step of extracting includes a sub-step of extracting from the message, as a modificative pictorial-symbol, at least one of the used pictorial-symbols in the message which is coincident with at least one of the at least one selected pictorial-symbol stored in the pictorial-symbol dictionary memory,

the method further comprising the steps of;

retrieving within the pictorial-symbol dictionary memory, at least one of the at least one intensity rate which corresponds to the at least one extracted modificative-pictorial-symbol; and

detecting from the message, as a depended pictorial-symbol, at least one of the used pictorial-symbols on which each of the at least one extracted modificative-pictorial-symbol depends linguistically,

wherein the step of calculating includes a sub-step of calculating the aggregate net opinion score for the message, by multiplying the opinion score for each of the at least one detected depended-pictorial-symbol, by the intensity rate for each of the at least one extracted modificative-pictorial-symbol, wherein the opinion score and the intensity rate are defined according to the correspondence stored in the pictorial-symbol dictionary memory.

7. The method according to claim 6, further comprising segmenting the message into morphemes, wherein the step of extracting includes a sub-step of extracting at least, one of the morphemes of the segmented message, which corresponds to at least one of the designated pictorial symbols stored in the pictorial-symbol dictionary memory.

8. The method according to claim 6, further comprising the steps of:

by the use of a literal-string dictionary memory storing a correspondence between designated literal-strings to be rated and scores of opinions expressed by the respective literal-strings, extracting from the message, at least one of the used literal-strings in the message which is coincident with at least one of the designated literal-strings stored in the literal-string dictionary memory; and

retrieving within the literal-string dictionary memory, at least one of the opinion scores which corresponds to the at least one extracted literal-sting,

wherein the step of calculating includes a sub-step of calculating the aggregate net opinion score for the message, based on a sum of a first aggregate opinion score for the at least one extracted pictorial-symbol, and a second aggregate opinion score for the at least one extracted literal-string, wherein the first aggregate opinion score is calculated with the at least one retrieved opinion score for pictorial symbol, and the second aggregate opinion score is calculated with the at least one retrieved opinion score for literal string.