Dialog data generating

An electronic device, comprising:

• one or more processors;
• a memory storing one or more programs configured to be executed by the one or more processors, the one or more programs including instructions for:
• obtaining a target dialog data template, wherein the target dialog data template comprises one or more target single-round dialog data templates, wherein each target single-round dialog data template comprises one or more keyword slots and related information about each keyword slot, and wherein the related information about each keyword slot comprises location information and attribute information;
• for each keyword slot, determining, from a keyword data set based on at least the attribute information of the keyword slot, a target keyword that matches the keyword slot, wherein the keyword data set comprises a plurality of keywords and attribute information of each keyword;
• for each target single-round dialog data template of the target dialog data template, filling the target single-round dialog data template with one or more target keywords based on the location information of the one or more keyword slots, to obtain target dialog data, and
• training a neural network model by using the target dialog data;
• wherein the target dialog data template comprises a plurality of single-round dialog data templates, and wherein the obtaining the target dialog data template comprises:
• sampling a dialog data set stored in the memory based on a dialog action label by using a preset rule to obtain a target dialog action label sequence comprising a plurality of target dialog action labels, wherein the dialog data set comprises a plurality of dialog action labels and a plurality of single-round dialog data templates corresponding to the plurality of the dialog action labels;
• determining, based on each target dialog action label in the target dialog action label sequence, a target single-round dialog data template corresponding to the target dialog action label;
• and
• building, based on the target dialog action label sequence and a plurality of target single-round dialog data templates corresponding to the target dialog action label sequence, the target dialog data template comprising the plurality of target single-round dialog data templates,
• and wherein, the sampling the dialog data set based on the dialog action label by using the preset rule comprises:
• for each target dialog action label in the target dialog action label sequence,
• determining respective sampling probabilities of the plurality of dialog action labels based at least on a preceding target dialog action label of the target dialog action label;
• and
• sampling the dialog data set based on the respective sampling probabilities of the plurality of dialog action labels, to obtain the target dialog action label.

A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to:

• obtaining a target dialog data template, wherein the target dialog data template comprises one or more target single-round dialog data templates, wherein each target single-round dialog data template comprises one or more keyword slots and related information about each keyword slot, and wherein the related information about each keyword slot comprises location information and attribute information;
• for each keyword slot, determining, from a keyword data set based on at least the attribute information of the keyword slot, a target keyword that matches the keyword slot, wherein the keyword data set comprises a plurality of keywords and attribute information of each keyword;
• for each target single-round dialog data template of the target dialog data template, filling the target single-round dialog data template with one or more target keywords based on the location information of the one or more keyword slots, to obtain target dialog data, and
• training a neural network model by using the target dialog data;
• wherein the target dialog data template comprises a plurality of single-round dialog data templates, and wherein the obtaining the target dialog data template comprises:
• sampling a dialog data set stored in a memory of the electronic device based on a dialog action label by using a preset rule to obtain a target dialog action label sequence comprising a plurality of target dialog action labels, wherein the dialog data set comprises a plurality of dialog action labels and a plurality of single-round dialog data templates corresponding to the plurality of the dialog action labels;
• determining, based on each target dialog action label in the target dialog action label sequence, a target single-round dialog data template corresponding to the target dialog action label;
• and
• building, based on the target dialog action label sequence and a plurality of target single-round dialog data templates corresponding to the target dialog action label sequence, the target dialog data template comprising the plurality of target single-round dialog data templates,
• and wherein, the sampling the dialog data set based on the dialog action label by using the preset rule comprises:
• for each target dialog action label in the target dialog action label sequence,
• determining respective sampling probabilities of the plurality of dialog action labels based at least on a preceding target dialog action label of the target dialog action label;
• and
• sampling the dialog data set based on the respective sampling probabilities of the plurality of dialog action labels, to obtain the target dialog action label.