US 12,235,881 B2
Vector acquisition method, vector acquisition apparatus, and recording medium
Koki Umehara, Kyoto (JP); Kiyotaka Kasubuchi, Kyoto (JP); Akiko Yoshida, Kyoto (JP); Manri Terada, Kyoto (JP); and Yuki Sumiya, Kyoto (JP)
Assigned to SCREEN HOLDINGS CO., LTD., Kyoto (JP)
Filed by SCREEN HOLDINGS CO., LTD., Kyoto (JP)
Filed on Feb. 11, 2022, as Appl. No. 17/669,625.
Claims priority of application No. 2021-041736 (JP), filed on Mar. 15, 2021.
Prior Publication US 2022/0292124 A1, Sep. 15, 2022
Int. Cl. G06F 16/335 (2019.01); G06F 16/334 (2025.01); G06F 16/355 (2025.01); G06F 40/247 (2020.01); G06F 40/53 (2020.01); G06N 20/00 (2019.01)
CPC G06F 16/3347 (2019.01) [G06F 16/355 (2019.01); G06F 40/247 (2020.01); G06F 40/53 (2020.01); G06N 20/00 (2019.01)] 9 Claims
OG exemplary drawing
 
8. A vector acquisition apparatus using a learned model holding initial vectors corresponding to respective words, and capable of outputting, for input of a piece of text including a plurality of words, adjusted vectors corresponding to the respective words in the piece of text, the initial vectors being vectors learned in advance, the adjusted vectors being vectors obtained by adjusting the initial vectors, and achieved by a central processing unit (CPU) executing a program, the vector acquisition apparatus comprising:
an input unit to input, into the learned model, at least one piece of text including at least two of a plurality of words obtained by dividing a compound, the compound being a word divisible into the plurality of words; and
an acquisition unit to acquire a compound vector corresponding to the compound using an adjusted vector output from the learned model and corresponding to at least one of the words obtained by dividing the compound in the input piece of text.