Method, apparatus, device and storage medium for information retrieval
According to embodiments of the disclosure, a method, an apparatus, a device and a storage medium for information retrieval are provided. In the method, input content for indicating a retrieval requirement is obtained from a target object; a set of retrieval elements are determined based on the input content, the set of retrieval elements comprising at least action elements for describing an event associated with a business object to be retrieved; and a set of retrieval results for the retrieval requirement are provided based on the set of retrieval elements and historical interaction information, the historical interaction information being generated based on a set of interaction events for at least one business component. Thereby, the embodiments of the disclosure can support natural language search of the business object by the user by utilizing the historical interaction information between the user and the business component.
1 . A method of information retrieval comprises:
obtaining, from a requester, input content for indicating a retrieval requirement;
determining a set of retrieval elements based on the input content, wherein the set of retrieval elements comprises at least action elements for describing an event associated with a business object to be retrieved; and
providing a set of retrieval results for the retrieval requirement based on the set of retrieval elements and historical interaction information, wherein the historical interaction information is generated based on a set of interaction events for at least one business component, and the business object is an object within the at least one business component,
wherein providing the set of retrieval results for the retrieval requirement comprises:
providing the set of retrieval results for the retrieval requirement based on vectors of the action elements and a reference vector set associated with a business object set, wherein the reference vector set is generated based on events associated with respective business objects, indicated by the historical interaction information.
2 . The method of claim 1 , wherein the historical interaction information is associated with a set of candidate business objects, and the set of retrieval results comprises at least one business object determined from the set of candidate business objects.
3 . The method of claim 1 , wherein the requester is a first requester, and the set of interaction events are associated with the first requester and/or at least one second requester.
4 . The method of claim 1 , wherein determining the set of retrieval elements based on the input content comprises:
providing the input content to a first model to determine whether the input content indicates a natural language search scene; and
in response to determining that the input content indicates the natural language search scene, determining the set of retrieval elements based on the input content.
5 . The method of claim 1 , wherein obtaining the input content for indicating the retrieval requirement comprises:
obtaining the input content of the requester received in a retrieval control; or
obtaining the input content input by the requester to the digital assistant in a conversation between the requester and the digital assistant.
6 . The method of claim 1 , wherein the set of retrieval elements further comprises at least one of:
a knowledge element for describing content of the business object;
a user element associated with the business object;
a time element associated with the business object;
a type element of the business object.
7 . The method of claim 1 , wherein determining the set of retrieval elements based on the input content comprises:
providing the input content to a second model; and
obtaining the set of retrieval elements determined by the second model based on the input content.
8 . The method of claim 7 , wherein the second model is trained based on a process comprising:
constructing a set of training samples based on a set of reference action elements and a set of reference knowledge elements; and
training the second model based on the set of training samples and the set of reference action elements.
9 . The method of claim 8 , wherein constructing the set of training input samples based on the set of reference action elements and the set of reference knowledge elements comprises:
generating an intermediate sample by combining a specific reference action element of the reference action elements and a corresponding reference knowledge element of the reference knowledge elements; and
processing the intermediate sample by using a third model, to obtain at least one training sample corresponding to the specific reference action element.
10 . The method of claim 9 , wherein the at least one training sample comprises:
a first training sample output by the third model; and/or
a second training sample, the second training sample being generated by deleting at least one limiting element in the first training sample.
11 . The method of claim 1 , wherein determining the set of retrieval elements based on the input content comprises:
obtaining association information of the requester, the association information indicating a set of associated objects associated with the requester, and
determining, based on the association information, the set of retrieval elements indicated by the input content.
12 . The method of claim 1 , wherein providing the set of retrieval results for the retrieval requirement comprises:
providing at least one retrieval result matching a part of retrieval elements in the set of retrieval elements.
13 . The method of claim 12 , further comprises:
providing matching information to the requester to indicate that the at least one result mismatches at least one retrieval element in the set of retrieval elements.
14 . The method of claim 1 , further comprises:
causing the set of retrieval results and the set of retrieval elements to be displayed in a result page, as a response to the retrieval requirement.
15 . An electronic device comprises:
at least one processing unit; and
at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform acts comprising:
obtaining, from a requester, input content for indicating a retrieval requirement;
determining a set of retrieval elements based on the input content, wherein the set of retrieval elements comprises at least action elements for describing an event associated with a business object to be retrieved; and
providing a set of retrieval results for the retrieval requirement based on the set of retrieval elements and historical interaction information, wherein the historical interaction information is generated based on a set of interaction events for at least one business component, and the business object is an object within the at least one business component,
wherein providing the set of retrieval results for the retrieval requirement comprises:
providing the set of retrieval results for the retrieval requirement based on vectors of the action elements and a reference vector set associated with a business object set, wherein the reference vector set is generated based on events associated with respective business objects, indicated by the historical interaction information.
16 . The electronic device of claim 15 , wherein the historical interaction information is associated with a set of candidate business objects, and the set of retrieval results comprises at least one business object determined from the set of candidate business objects.
17 . The electronic device of claim 15 , wherein the requester is a first requester, and the set of interaction events are associated with the first requester and/or at least one second requester.
18 . The electronic device of claim 15 , wherein determining the set of retrieval elements based on the input content comprises:
providing the input content to a first model to determine whether the input content indicates a natural language search scene; and
in response to determining that the input content indicates the natural language retrieval scene, determining the set of retrieval elements based on the input content.
19 . A non-transitory computer-readable storage medium having stored thereon a computer program executable by a processor to implement acts comprising:
obtaining, from a requester, input content for indicating a retrieval requirement;
determining a set of retrieval elements based on the input content, wherein the set of retrieval elements comprises at least action elements for describing an event associated with a business object to be retrieved; and
providing a set of retrieval results for the retrieval requirement based on the set of retrieval elements and historical interaction information, wherein the historical interaction information is generated based on a set of interaction events for at least one business component, and the business object is an object within the at least one business component,
wherein providing the set of retrieval results for the retrieval requirement comprises:
providing the set of retrieval results for the retrieval requirement based on vectors of the action elements and a reference vector set associated with a business object set, wherein the reference vector set is generated based on events associated with respective business objects, indicated by the historical interaction information.