CN113360537B - Information query method, device, electronic equipment and medium - Google Patents
Information query method, device, electronic equipment and medium Download PDFInfo
- Publication number
- CN113360537B CN113360537B CN202110625927.4A CN202110625927A CN113360537B CN 113360537 B CN113360537 B CN 113360537B CN 202110625927 A CN202110625927 A CN 202110625927A CN 113360537 B CN113360537 B CN 113360537B
- Authority
- CN
- China
- Prior art keywords
- query
- candidate
- segment
- word
- candidate query
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 239000012634 fragment Substances 0.000 claims abstract description 91
- 238000013519 translation Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000004590 computer program Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 210000000744 eyelid Anatomy 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000006467 substitution reaction Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010845 search algorithm Methods 0.000 description 3
- 244000017020 Ipomoea batatas Species 0.000 description 2
- 235000002678 Ipomoea batatas Nutrition 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000001356 surgical procedure Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
- G06F16/2246—Trees, e.g. B+trees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2452—Query translation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure provides an information query method, an information query device, electronic equipment and a medium, relates to the technical field of computers, and particularly relates to the technical fields of intelligent search technology, cloud computing and cloud service. The specific implementation scheme is as follows: determining target query words matched with the current query segment according to the incidence relation between the current query segment input by the user and the candidate query segment and the candidate query words; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment; and carrying out information query according to the target query words. The method and the device improve the recall efficiency of the query words, and further improve the information query efficiency.
Description
Technical Field
The disclosure relates to the technical field of computers, in particular to the technical fields of intelligent search technology, cloud computing and cloud service, and particularly relates to a cluster access method, a cluster access device, electronic equipment and a medium.
Background
The automatic query term completion function can greatly improve the search efficiency of a user, and is generally used in the input frame scene of a search engine or an application program, and automatically completes complete query terms which the user may want to search according to prefixes which the user has input.
Automatic completion of query terms is currently typically accomplished based on a prefix tree of constructed query terms.
Disclosure of Invention
The disclosure provides a method, a device, electronic equipment and a medium for inquiring information according to a current inquiry segment input by a user.
According to an aspect of the present disclosure, there is provided an information query method including:
determining a target query word matched with a current query segment according to the association relation between the current query segment input by a user and the candidate query segment and the candidate query word; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment;
and inquiring information according to the target inquiry words.
According to another aspect of the present disclosure, there is provided an information inquiry apparatus including:
the target query word determining module is used for determining target query words matched with the current query segment according to the incidence relation between the current query segment input by the user and the candidate query segment and the candidate query words; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment;
and the information query module is used for carrying out information query according to the target query words.
According to another aspect of the present disclosure, there is provided an electronic device including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of the present disclosure.
According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the method of any one of the present disclosure.
According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements a method according to any of the present disclosure.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 is a flow chart of a method of querying information disclosed in accordance with an embodiment of the present disclosure;
FIG. 2A is a flow chart of a method of querying information disclosed in accordance with an embodiment of the present disclosure;
FIG. 2B is a schematic diagram of generating candidate query snippets in accordance with an embodiment of the present disclosure;
fig. 3 is a schematic structural view of an information query apparatus according to an embodiment of the present disclosure;
fig. 4 is a block diagram of an electronic device for implementing the information query method disclosed in the embodiments of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Applicant has found during development that existing automatic completion of query terms is typically achieved from a tree of prefixes of the constructed query terms, e.g., for query term "download 54321", the constructed tree of prefixes optionally including the following prefixes: "download", "download 5", "download 54", "download 543", "download 5432", "download 54321", … …. When the user inputs the current query fragment in the information query interface, if the current query fragment is completely matched with any prefix in any prefix tree, recall the query word corresponding to the prefix tree, for example, the user inputs "download 5", and then recall "download 54321" according to the constructed prefix tree.
However, in the existing query term recall method, the query term can be recalled only by completely matching the current query segment with the prefix in the prefix tree, if the current query segment input by the user is not standard, the query term can be recalled only by clarifying the query intention of the user for many times, so that the recall efficiency of each query term is lower. For example, the current query fragment has wrongly written words, such as "download" to "frighten" the load; for another example, the current query segment has a mixed Chinese character and pinyin, such as "download" to "next zai", etc. And in the case of lower recall efficiency of each query term, the information query efficiency is further lower.
Fig. 1 is a flowchart of an information query method according to an embodiment of the present disclosure, which may be applicable to a case of performing information query according to a current query segment input by a user. The method of the embodiment can be executed by the information query device disclosed by the embodiment of the disclosure, and the device can be realized by software and/or hardware and can be integrated on any electronic equipment with computing capability.
As shown in fig. 1, the information query method disclosed in this embodiment may include:
s101, determining target query words matched with a current query segment according to the association relation between the current query segment input by a user and the candidate query segment and the candidate query words; the candidate query segment is determined according to the candidate query word, the historical query segment and the historical query word recalled by the historical query segment.
The current query segment represents information which is currently input by a user in an input box of the information query page. The candidate query terms are terms which are obtained in advance and are set to be completed, for example, hot terms are crawled in the internet based on a crawler technology, and the crawled hot terms are used as candidate query terms in the embodiment; for another example, words are obtained from a Xinhua dictionary or other more formal literature materials, and the obtained words are used as candidate query words in the embodiment; for another example, according to the exposure requirement for some terms in the actual service, the terms are artificially created to serve as candidate query terms in this embodiment, and this embodiment does not limit any method for obtaining the candidate query terms. The candidate query snippets are generated from each candidate query term, i.e., each candidate query term contains at least one candidate query snippet. And, in addition, the processing unit,
the association relation between each candidate query word and the candidate query fragments contained in the candidate query word is pre-constructed, namely, which candidate query fragments are contained in the candidate query word can be determined according to any candidate query word, and the corresponding candidate query word can be determined according to any candidate query fragment.
In one embodiment, a historical query segment input by a user at any historical moment is obtained from a query click log, and a historical query word finally recalled according to the historical query segment is obtained, wherein the historical query segment can be divided into a prefix matching segment and a non-prefix matching segment.
The prefix matching segment represents a query segment which can be completely matched with any prefix in any prefix tree, and correspondingly, the history query words recalled by the prefix matching segment are query words which the prefix tree belongs to.
The non-prefix matching segment represents a query segment which cannot be completely matched with any prefix in any prefix tree, and the query segment needs to be transformed for the non-prefix matching segment and then recalled by a query word, wherein the transformation mode comprises at least one of the following modes: 1) A suffix tree is constructed, for example, query term recall is performed with the suffix "12" of the non-prefix matching segment "install 12" as a new query segment, and then spliced to the rear of "install 12" to make up the complete query term. 2) Synonyms are mined, i.e. alternative synonyms are used to match the prefix tree, e.g. the set of synonyms < install, download > is mined, and for non-prefix matching segments "install 12", after "download" replaces "install", query recall is performed with "download 12". 3) Training a semantic model, constructing a semantic vector for the non-prefix matching segment, and recalling in a manner of performing neighbor retrieval on the semantic vector of the non-prefix matching segment and the semantic vector of the query word. 4) And adopting a machine translation model, namely carrying out machine translation on the non-prefix matching segments through the machine translation model to obtain the query words.
After the history inquiry fragments and the history inquiry words recalled by the history inquiry fragments at any time are obtained, the corresponding relation between the history inquiry fragments and the history inquiry words is learned by adopting a machine learning mode, namely, which history inquiry fragments can be recalled by one history inquiry word is learned. After machine learning is completed, predicting which candidate query fragments can be recalled by each candidate query word through the corresponding relation of the learning completion, and obtaining the candidate query fragments corresponding to each candidate query word. And finally, matching the current query segment input by the user with the candidate query segment corresponding to each candidate query word, and determining the target query word matched with the current query segment from the candidate query words according to the matching result and the association relation between the candidate query segment and the candidate query word.
According to the method, the target query word matched with the current query segment is determined according to the association relation between the current query segment input by the user and the candidate query segment and the candidate query word, so that the effect of determining the target query word according to the current query segment input by the user is realized, and a foundation is laid for information query according to the target query word.
S102, information inquiry is carried out according to the target inquiry words.
In one embodiment, the search engine first performs word segmentation on the target query word and removes stop words. The search engine program then locates the relevant web pages containing the target query terms from the index database, sorts the web pages, and returns the web pages to the information query page in a format. Factors that affect web page ranking are numerous, including but not limited to: the target query word commonly used degree, the word frequency and density of the target query word, the position and form of the target query word, the distance and link analysis of the target query word, the page weight and the like.
According to the method and the device, the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment, and then according to the current query fragment input by a user, the association relation between the candidate query fragment and the candidate query word is determined, the target query word matched with the current query fragment is finally queried according to the target query word, and because the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment, the candidate query fragment is more suitable for the query fragment input by the user wanting to recall the candidate query word in an actual scene, and further the target query word corresponding to the current query fragment can be quickly matched according to the candidate query fragment without changing the current query fragment, so that the recall efficiency of the query word is greatly improved, and the information query efficiency is improved.
Fig. 2A is a flowchart of an information query method according to an embodiment of the present disclosure, which is further optimized and expanded based on the above technical solution, and may be combined with the above various alternative embodiments.
As shown in fig. 2A, the information query method disclosed in this embodiment may include:
s201, taking the history query words recalled by the history query segments as input, taking the history query segments as output, and training the model to be trained to obtain a candidate query segment generation model.
In one embodiment, a preset number of historical query segments are obtained from a query log, the historical query segments are screened, the historical query segments with byte length smaller than the preset length are removed, and the preset length can be 2. Correspondingly, the history query words which are finally recalled by the selected history query fragments are obtained, and then any history query fragment and at least one history query word which corresponds to the history query fragment are used as a group of training data. After all training data are collected, the historical query words are used as input of a model to be trained, the historical query fragments corresponding to the historical query words are used as output, the model to be trained is trained, and a candidate query fragment generation model is generated.
Alternatively, the present embodiment may artificially construct some training data, including but not limited to: 1) The synonymous query is constructed, for example, the query "price for double eyelid surgery" is derived to obtain "cost for double eyelid surgery". 2) Synonym substitutions, such as english case substitutions, chinese synonym substitutions, and/or number case substitutions, among others. 3) Based on the replacement of Chinese character strokes, for example, the query segment of "sweet potato nine" is constructed according to the query word of "sweet potato ball", etc.
Through artificial constitution training data, can enrich the variety of training data, and then make model training result better.
Optionally, the model to be trained is a neural network translation model.
In one embodiment, the historical query word is used as an end source of the neural network translation model, the historical query fragment is used as a target end of the neural network translation model, namely, the historical query word is input, and the historical query fragment is output. And constructing an enabling dictionary based on the word model, and training by adopting a transducer model to obtain a neural network translation model. In this embodiment, the neural network translation model may be selected from an LSTM model, an SMT model, or the like, in addition to the transducer model, and the specific type of the neural network translation model is not limited in this embodiment.
By setting the model to be trained as the neural network translation model, better effects can be generated on the character conversion scene due to the model characteristics of the neural network translation model, so that the candidate query fragments obtained later are more accurate.
S202, taking the candidate query words as input of a candidate query segment generation model, and obtaining the candidate query segments.
In one embodiment, a preset candidate query word is input into a trained candidate query segment generation model, the candidate query segment generation model predicts the candidate query word according to the relationship between the learned query word and the query segment, predicts which query segments the candidate query word can recall from, and outputs at least one query segment with higher probability as the candidate query segment corresponding to the candidate query word according to the calculated probability.
Fig. 2B is a schematic diagram of generating a candidate query segment according to an embodiment of the present disclosure, as shown in fig. 2B, 20 is a candidate query segment generation model, in this embodiment, the candidate query segment generation model 20 is selected as a transducer model, 21 is a candidate query word, taking "double eyelid operation price" as an example, and 22 is a candidate query segment corresponding to the candidate query word 21, taking "double eyelid sho" as an example. It can be seen that, based on the relationship between the query word and the query segment learned by the candidate query segment generation model 20, for the candidate query word "double eyelid operation price" 21, the candidate query segment generation model 20 predicts that the user will input the query word "double eyelid sho", i.e., the candidate query segment "double eyelid sho"22, to recall the candidate query word "double eyelid operation price" 21.
Optionally, the method further comprises obtaining candidate query segments by:
and carrying out character splitting on the candidate query words, and obtaining the candidate query fragments according to character splitting results.
In one embodiment, the characters included in each candidate query term are split, the split character strings are combined, and then each character string is combined to be used as a candidate query segment of the candidate query term.
The candidate query words are subjected to character splitting, and candidate query fragments are obtained according to character splitting results, so that the candidate query fragments comprise query fragments predicted by a candidate query fragment generation model and query fragments related to the characters of the candidate query words, and recall of the query words can be directly performed no matter whether the current query fragments input by a user are standard or not, and the recall efficiency of the query words is improved.
Optionally, after obtaining the candidate query segment, the method further includes:
and expanding the candidate query fragments by adopting a heuristic search algorithm.
In one embodiment, a heuristic search algorithm is used to expand the related terms of each candidate query segment, so as to increase the coverage range of the candidate query segment.
By expanding the candidate query fragments by adopting a heuristic search algorithm, the diversity and coverage of the candidate query fragments are improved, and the recall rate of query words is further improved.
S203, determining target query fragments matched with the current query fragments from the candidate query fragments according to the current query fragments input by the user.
In one embodiment, a current query segment input by a user in an input box of an information query interface is obtained, the current query segment is matched in candidate query segments, and the candidate query segment matched with the current query segment is used as a target query segment.
Optionally, the current query segment includes a character query segment and/or a phoneme query segment.
The character query segment includes but is not limited to characters, such as Chinese, english, japanese, etc., numbers, punctuation, etc. The phoneme query fragments include, but are not limited to, pinyin and phonetic symbols, etc.
For example, assuming that the current query segment is "today's tie qi", if "today's tie qi" is included in the candidate query segment, then "today's tie qi" is taken as the target query segment.
By setting the current query segment to comprise a character query segment and/or a phoneme query segment, the application range of information query is improved.
S204, determining the target query word from the candidate query words according to the target query segment and the association relation between the candidate query segment and the candidate query word.
In one embodiment, at least one candidate query term associated with the target query segment is determined based on an association between the candidate query segment and the candidate query term, and the at least one candidate query term associated with the target query segment is taken as the target query term.
For example, assuming that the target query segment is "today's tie qi", and candidate query words corresponding to "today's tie qi" in the preset association are "how today's weather", "how today's weather" and "how today's weather is hot", at least one of "how today's weather", "how today's weather" and "how today's weather is hot" is taken as the target query word.
S205, information inquiry is carried out according to the target inquiry words.
In one embodiment, if the number of target query terms is at least two, the user may select from at least two target query terms, and then perform information query according to the final target query term selected by the user.
In another embodiment, if the number of the target query words is at least two, information query is performed according to each target query word, and finally the information query results are summarized and presented to the user.
According to the method, the candidate query terms are used as the input of the candidate query fragment generation model, the candidate query fragments are obtained, the candidate query fragment generation model is obtained by taking the historical query terms as the input and taking the historical query fragments as the output, and training the model to be trained; the target query fragment matched with the current query fragment is determined from the candidate query fragments, and the target query word is determined from the candidate query words according to the target query fragment and the association relation between the candidate query fragment and the candidate query word, so that the effect of matching and determining the corresponding target query word according to the current query fragment input by the user is realized, and a foundation is laid for follow-up information query.
Fig. 3 is a schematic structural diagram of an information query apparatus according to an embodiment of the present disclosure, which may be suitable for a case of performing information query according to a current query segment input by a user. The device of the embodiment can be implemented by software and/or hardware, and can be integrated on any electronic equipment with computing capability.
As shown in fig. 3, the information query apparatus 30 disclosed in this embodiment may include a target query term determining module 31 and an information query module 32, where:
the target query term determining module 31 is configured to determine a target query term matched with a current query segment according to an association relationship between the current query segment input by a user and the candidate query segment and the candidate query term; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment;
and the information query module 32 is used for querying information according to the target query words.
Optionally, the current query segment includes a character query segment and/or a phoneme query segment.
Optionally, the apparatus further includes a first candidate query segment generation module, specifically configured to:
taking the candidate query words as the input of a candidate query segment generation model to obtain the candidate query segments;
the candidate query segment generation model is obtained by training a model to be trained by taking the historical query word as input and taking the historical query segment as output.
Optionally, the model to be trained is a neural network translation model.
Optionally, the apparatus further includes a second candidate query segment generation module, specifically configured to:
and carrying out character splitting on the candidate query words, and obtaining the candidate query fragments according to character splitting results.
Optionally, the target query term determining module 31 is specifically configured to:
determining target query fragments matched with the current query fragment from the candidate query fragments;
and determining the target query word from the candidate query words according to the target query segment and the association relation between the candidate query segment and the candidate query word.
The information query device 30 disclosed in the embodiments of the present disclosure may execute the information query method disclosed in the embodiments of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method. Reference is made to the description of any method embodiment of the disclosure for details not explicitly described in this embodiment.
…
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
Fig. 4 illustrates a schematic block diagram of an example electronic device 400 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 4, the apparatus 400 includes a computing unit 401 that can perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM) 402 or a computer program loaded from a storage unit 408 into a Random Access Memory (RAM) 403. In RAM 403, various programs and data required for the operation of device 400 may also be stored. The computing unit 401, ROM 402, and RAM 403 are connected to each other by a bus 404. An input/output (I/O) interface 405 is also connected to bus 404.
Various components in device 400 are connected to I/O interface 405, including: an input unit 406 such as a keyboard, a mouse, etc.; an output unit 407 such as various types of displays, speakers, and the like; a storage unit 408, such as a magnetic disk, optical disk, etc.; and a communication unit 409 such as a network card, modem, wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The computing unit 401 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 401 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the respective methods and processes described above, such as the information inquiry method. For example, in some embodiments, the information query method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 400 via the ROM 402 and/or the communication unit 409. When the computer program is loaded into RAM 403 and executed by computing unit 401, one or more steps of the information inquiry method described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the information query method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (10)
1. An information query method, comprising:
the candidate query words are used as the input of a candidate query segment generation model, and candidate query segments are obtained; the candidate query segment generation model is obtained by taking a historical query word as input and a historical query segment as output, and training a model to be trained;
determining a target query word matched with a current query segment according to the association relation between the current query segment input by a user and the candidate query segment and the candidate query word; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment;
inquiring information according to the target inquiry words;
according to the association relation between the current query segment input by the user and the candidate query segment and the candidate query word, determining the target query word matched with the current query segment comprises the following steps:
determining target query fragments matched with the current query fragment from the candidate query fragments;
and determining the target query word from the candidate query words according to the target query segment and the association relation between the candidate query segment and the candidate query word.
2. The method of claim 1, wherein the current query segment comprises a character query segment and/or a phoneme query segment.
3. The method of claim 1, wherein the model to be trained is a neural network translation model.
4. The method of claim 1, wherein prior to determining the target query term matching the current query segment based on the association between the current query segment entered by the user and the candidate query segment and the candidate query term, further comprising:
and carrying out character splitting on the candidate query words, and obtaining the candidate query fragments according to character splitting results.
5. An information query apparatus, comprising:
the first candidate query fragment generation module is specifically configured to:
taking the candidate query words as the input of a candidate query segment generation model to obtain the candidate query segments; the candidate query segment generation model is obtained by taking a historical query word as input and a historical query segment as output, and training a model to be trained;
the target query word determining module is used for determining target query words matched with the current query segment according to the incidence relation between the current query segment input by the user and the candidate query segment and the candidate query words; the candidate query fragment is determined according to the candidate query word, the historical query fragment and the historical query word recalled by the historical query fragment;
the information inquiry module is used for inquiring information according to the target inquiry words;
the target query term determining module is specifically configured to:
determining target query fragments matched with the current query fragment from the candidate query fragments;
and determining the target query word from the candidate query words according to the target query segment and the association relation between the candidate query segment and the candidate query word.
6. The apparatus of claim 5, wherein the current query segment comprises a character query segment and/or a phoneme query segment.
7. The apparatus of claim 5, wherein the model to be trained is a neural network translation model.
8. The apparatus of claim 5, wherein the apparatus further comprises a second candidate query fragment generation module, specifically configured to:
and carrying out character splitting on the candidate query words, and obtaining the candidate query fragments according to character splitting results.
9. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-4.
10. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110625927.4A CN113360537B (en) | 2021-06-04 | 2021-06-04 | Information query method, device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110625927.4A CN113360537B (en) | 2021-06-04 | 2021-06-04 | Information query method, device, electronic equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113360537A CN113360537A (en) | 2021-09-07 |
| CN113360537B true CN113360537B (en) | 2024-01-12 |
Family
ID=77532365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110625927.4A Active CN113360537B (en) | 2021-06-04 | 2021-06-04 | Information query method, device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113360537B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107291914A (en) * | 2017-06-27 | 2017-10-24 | 达而观信息科技（上海）有限公司 | A kind of method and system for generating search engine inquiry expansion word |
| CN108304444A (en) * | 2017-11-30 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Information query method and device |
| CN108733766A (en) * | 2018-04-17 | 2018-11-02 | 腾讯科技（深圳）有限公司 | A kind of data query method, apparatus and readable medium |
| CN109145213A (en) * | 2018-08-22 | 2019-01-04 | 清华大学 | Inquiry recommended method and device based on historical information |
| CN111125310A (en) * | 2019-12-24 | 2020-05-08 | 北京百度网讯科技有限公司 | Voice query method and device |
| CN111414746A (en) * | 2020-04-10 | 2020-07-14 | 中国建设银行股份有限公司 | Matching statement determination method, device, equipment and storage medium |
| CN111767477A (en) * | 2020-06-19 | 2020-10-13 | 北京百度网讯科技有限公司 | Retrieval method, retrieval device, electronic equipment and storage medium |
| EP3743827A1 (en) * | 2019-02-01 | 2020-12-02 | Google LLC | Training image and text embedding models |
-
2021
- 2021-06-04 CN CN202110625927.4A patent/CN113360537B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107291914A (en) * | 2017-06-27 | 2017-10-24 | 达而观信息科技（上海）有限公司 | A kind of method and system for generating search engine inquiry expansion word |
| CN108304444A (en) * | 2017-11-30 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Information query method and device |
| CN108733766A (en) * | 2018-04-17 | 2018-11-02 | 腾讯科技（深圳）有限公司 | A kind of data query method, apparatus and readable medium |
| CN109145213A (en) * | 2018-08-22 | 2019-01-04 | 清华大学 | Inquiry recommended method and device based on historical information |
| EP3743827A1 (en) * | 2019-02-01 | 2020-12-02 | Google LLC | Training image and text embedding models |
| CN111125310A (en) * | 2019-12-24 | 2020-05-08 | 北京百度网讯科技有限公司 | Voice query method and device |
| CN111414746A (en) * | 2020-04-10 | 2020-07-14 | 中国建设银行股份有限公司 | Matching statement determination method, device, equipment and storage medium |
| CN111767477A (en) * | 2020-06-19 | 2020-10-13 | 北京百度网讯科技有限公司 | Retrieval method, retrieval device, electronic equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 基于Word2vec的语义查询扩展方法;章露露;吕晓伟;;软件导刊(09);全文 * |
| 机器学习的查询扩展在博客检索中的应用;王秉卿;张奇;吴立德;黄萱菁;;中文信息学报(06);全文 * |
| 查询推荐研究综述;张晓娟;彭琳;李倩;;情报学报(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113360537A (en) | 2021-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111563208B (en) | Method and device for identifying intention and computer readable storage medium | |
| CN107301170B (en) | Method and device for segmenting sentences based on artificial intelligence | |
| KR102464248B1 (en) | Method, apparatus, electronic device, and storage medium for extracting spo triples | |
| CN103268313B (en) | A kind of semantic analytic method of natural language and device | |
| US11907671B2 (en) | Role labeling method, electronic device and storage medium | |
| CN113553414B (en) | Intelligent dialogue method, intelligent dialogue device, electronic equipment and storage medium | |
| EP3879416A1 (en) | Query auto-completion method, apparatus and device, and computer storage medium | |
| CN112541125B (en) | Sequence annotation model training method and device and electronic equipment | |
| CN112925900B (en) | Search information processing method, device, equipment and storage medium | |
| CN112506359B (en) | Method and device for providing candidate long sentences in input method and electronic equipment | |
| JP7241122B2 (en) | Smart response method and device, electronic device, storage medium and computer program | |
| CN113033194B (en) | Training method, device, equipment and storage medium for semantic representation graph model | |
| CN112560425B (en) | Template generation method and device, electronic equipment and storage medium | |
| CN113360537B (en) | Information query method, device, electronic equipment and medium | |
| CN113807091B (en) | Word mining method and device, electronic equipment and readable storage medium | |
| CN113204613B (en) | Address generation method, device, equipment and storage medium | |
| CN112948573B (en) | Text label extraction method, device, equipment and computer storage medium | |
| CN111797205B (en) | Vocabulary retrieval method and device, electronic equipment and storage medium | |
| CN114417862A (en) | Text matching method, and training method and device of text matching model | |
| CN112905917A (en) | Inner chain generation method, model training method, related device and electronic equipment | |
| CN113377921B (en) | Method, device, electronic equipment and medium for matching information | |
| CN113377922B (en) | Method, device, electronic equipment and medium for matching information | |
| CN112905777B (en) | Extended query recommendation method and device, electronic equipment and storage medium | |
| CN114896993B (en) | Translation model generation method and device, electronic equipment and storage medium | |
| CN113033196B (en) | Word segmentation method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |