CN105760399A

CN105760399A - Data retrieval method and device

Info

Publication number: CN105760399A
Application number: CN201410799739.3A
Authority: CN
Inventors: 赵荣乾
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd; Huawei Software Technologies Co Ltd
Priority date: 2014-12-19
Filing date: 2014-12-19
Publication date: 2016-07-13

Abstract

The embodiment of the invention discloses a data retrieval method and device and relates to the technical field of communication.The data retrieval method and device are used for improving retrieval result accuracy and retrieval efficiency.According to the specific scheme, the data retrieval method comprises the steps that a call center server receives a voice message sent by a user terminal, and a text message obtained by carrying out text conversion on the voice message is obtained; a first segmentation set recognized by carrying out text analysis on the text message is obtained, wherein the first segmentation set comprises at least one text segmentation; a keyword index list of retrieval data stored in a knowledge database is searched for retrieval keywords matched with the text segmentations in the first segmentation set, wherein the keyword index list contains at least one keyword index entry, and the keyword index entry contains retrieval keywords and identities of retrieval data corresponding to the retrieval keywords; the retrieval data indicated by the identifies of the retrieval data corresponding to the found retrieval keywords is retrieved in the knowledge database.The data retrieval method and device are used in the data retrieval process.

Description

A kind of data retrieval method and equipment

Technical field

The present invention relates to communication technical field, particularly relate to a kind of data retrieval method and equipment.

Background technology

Along with information-based high speed development, the information reservation amount in industry-by-industry or field also gets more and more.Such as, provide the user the operator of communication service in order to provide a user with comparatively comprehensively, fast and accurately business consultation and charge querying etc. service, the contact staff constructing these telecom operators can carry out the knowledge data base of data retrieval.

Wherein, contact staff after connected the voice communication with user by Call Center Server, then can identify search key according to the business experience of individual and technical ability from user speech information；Then by search key Input knowledge data base, knowledge data base match the information mated with search key as retrieval data, and show retrieval data to contact staff；Contact staff also needs to retrieval extracting data that the business experience according to individual and technical ability shows from knowledge data base and the higher retrieval data of user speech information relevance, and make answer according to the retrieval data extracted for user, or by the retrieval data feedback that extracts to user.

But, when in above-mentioned data retrieval process, contact staff retrieves from knowledge data base, to quickly retrieve matching degree between user speech information and the higher retrieval data of the degree of association, then needing contact staff to possess the business experience of abundant individual and the business skill of higher individual, the accuracy of retrieval result is too dependent on business experience and the technical ability of the individual of contact staff；Further, contact staff inevitably owing to some retrieval errors occur in carelessness or general idea, then can affect accuracy and the recall precision of retrieval in retrieving, causes that the accuracy of retrieval result and recall precision are relatively low.

Summary of the invention

Embodiments of the invention provide a kind of data retrieval method and equipment, in order to improve the accuracy of retrieval result, and improve recall precision.

For reaching above-mentioned purpose, embodiments of the invention adopt the following technical scheme that

The first aspect of the embodiment of the present invention, it is provided that a kind of data retrieval method, including:

Call Center Server receives the voice messaging that user terminal sends, and described voice messaging is carried out the text message that text is converted to by acquisition；

Described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, and comprises at least one text participle in described first participle set；

In the key word index list of the retrieval data that described Call Center Server stores from knowledge data base, search the search key mated with the text participle in described first participle set, described key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Described Call Center Server from described knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

In conjunction with first aspect, in the implementation that the first is possible, in key word index item in the key word index list of the retrieval data stored from knowledge data base at described Call Center Server, before searching the search key mated with the text participle in described first participle set, described method also includes:

Described Call Center Server reads the first retrieval data of storage in described knowledge data base, and described first retrieval data are any one retrieval data of storage in described knowledge data base；

Described Call Center Server obtains and described first retrieval data is carried out the second participle set that text analyzing identifies, and comprises at least one text participle in described second participle set；

Described Call Center Server, according to the text participle in described second participle set, generates the search key of described first retrieval data；

Described Call Center Server generates a described key word index item according to the mark of the described first search key retrieving data and described first retrieval data, generates the described key word index list comprising at least one described key word index item.

In conjunction with first aspect, in the implementation that the second is possible, in key word index item in the key word index list of the retrieval data stored from described knowledge data base at described Call Center Server, before searching the search key mated with the text participle in described first participle set, described method also includes:

Receive the described key word index list that text analyzing server sends.

In conjunction with any one in the implementation that first aspect, implementation that the first is possible and the second are possible, in the implementation that the third is possible, described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

Described text message is carried out text analyzing by described Call Center Server, identifies the 3rd participle set, comprises at least one text participle in described 3rd participle set；

Described Call Center Server filters the meaningless text participle comprised in described 3rd participle set, obtains described first participle set, and described meaningless text participle includes: exclamation and modal particle.

In conjunction with any one in the implementation that first aspect, implementation that the first is possible and the second are possible, in the 4th kind of possible implementation, described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

Described Call Center Server filters the first text participle comprised in described 3rd participle set, obtaining described first participle set, described first text participle is text participle all unmatched with any one search key in described key word index list in described 3rd participle set.

In conjunction with the implementation that first aspect or the second are possible, in the 5th kind of possible implementation, described voice messaging is made up of at least one sound bite being dimension with sentence, and described text message is made up of at least one text fragments；

Described Call Center Server receives the voice messaging that user terminal sends, and described voice messaging is carried out the text message that text is converted to by acquisition, including:

Described Call Center Server receives the described sound bite that described user terminal sends；

Described Call Center Server, after receiving a described sound bite, sends described sound bite to voice analysis server, so that described sound bite is converted to described text fragments by described voice analysis server；

Described Call Center Server receives the described text fragments that described voice analysis server sends.

In conjunction with the implementation that first aspect or the second are possible, in the 6th kind of possible implementation, described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

Described Call Center Server sends described text message to described text analyzing server, so that described text message is carried out text analyzing by described text analyzing server, identifies described first participle set；

Described Call Center Server receives the described first participle set that described text analyzing server sends.

The second aspect of the embodiment of the present invention, it is provided that a kind of data retrieval method, including:

Text analyzing server receives the text message that Call Center Server sends, and described text message is that described Call Center Server is converted to according to voice messaging；

Described text message is carried out text analyzing by described text analyzing server, identifies described first participle set, comprises at least one text participle in described first participle set；

Described text analyzing server sends described first participle set to described Call Center Server, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base.

In conjunction with second aspect, in the implementation that the first is possible, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Described method also includes:

Described text analyzing server reads the first retrieval data of storage in described knowledge data base, and described first retrieval data are any one retrieval data of storage in described knowledge data base；

Described text analyzing server is retrieved data to described first and is carried out text analyzing, identifies the second participle set, comprises at least one text participle in described second participle set；

Described text analyzing server, according to the text participle in described second participle set, generates the search key of described first retrieval data；

Described text analyzing server generates a described key word index item according to the mark of the described first search key retrieving data and described first retrieval data, generates the described key word index list comprising at least one described key word index item；

Described text analyzing server sends described key word index list to described Call Center Server.

The third aspect of the embodiment of the present invention, it is provided that a kind of Call Center Server, including:

Receive unit, for receiving the voice messaging that user terminal sends；

Text conversion units, carries out, for obtaining the described voice messaging that described reception unit is received, the text message that text is converted to；

Text analysis unit, carries out, for obtaining the described text message that described text conversion units is obtained, the first participle set that text analyzing identifies, comprises at least one text participle in described first participle set；

Keyword search unit, for in the key word index list of the retrieval data of storage from knowledge data base, search the search key that the text participle in the described first participle set obtained with described text analysis unit mates, described key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Retrieval unit, for, from described knowledge data base, retrieving the retrieval data that the mark of the data of the retrieval corresponding to search key that described keyword search unit finds is indicated.

In conjunction with the third aspect, in the implementation that the first is possible, described Call Center Server, it is also possible to including:

Read unit, for in the key word index item in the key word index list of the retrieval data stored from described knowledge data base at described keyword search unit, before searching the search key mated with the text participle in described first participle set, reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base；

Described text analysis unit, is additionally operable to obtain the described first retrieval data that described reading unit is read and carries out the second participle set that text analyzing identifies, comprise at least one text participle in described second participle set；

Generate unit, for the text participle in the described second participle set that obtains according to described text analysis unit, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item.

In conjunction with the third aspect, in the implementation that the second is possible, described reception unit, it is additionally operable in the key word index item in the key word index list of retrieval data stored from described knowledge data base at described keyword search unit, before searching the search key mated with the text participle in described first participle set, receive the described key word index list that text analyzing server sends.

In conjunction with any one in the implementation that the third aspect, implementation that the first is possible and the second are possible, in the implementation that the third is possible, described text analysis unit, including:

Text analysis model, for described text message is carried out text analyzing, identifies the 3rd participle set, comprises at least one text participle in described 3rd participle set；

First filtering module, for filtering the meaningless text participle comprised in the described 3rd participle set that described text analysis model identifies, obtains described first participle set, and described meaningless text participle includes: exclamation and modal particle.

In conjunction with any one in the implementation that the third aspect, implementation that the first is possible and the second are possible, in the 4th kind of possible implementation, described text analysis unit, including:

Second filtering module, for filtering the first text participle comprised in the described 3rd participle set that described text analysis model identifies, obtaining described first participle set, described first text participle is text participle all unmatched with any one search key in described key word index list in described 3rd participle set.

In conjunction with the implementation that the third aspect or the second are possible, in the 5th kind of possible implementation, described voice messaging is made up of at least one sound bite being dimension with sentence, and described text message is made up of at least one text fragments；

Described reception unit, specifically for receiving the described sound bite that described user terminal sends；

Described text conversion units, specifically for, after described reception unit receives a described sound bite, sending described sound bite to voice analysis server, so that described sound bite is converted to described text fragments by described voice analysis server；Receive the described text fragments that described voice analysis server sends.

In conjunction with the implementation that the third aspect or the second are possible, in the 6th kind of possible implementation, described text analysis unit, specifically for sending described text message to described text analyzing server, so that described text message is carried out text analyzing by described text analyzing server, identify described first participle set；Receive the described first participle set that described text analyzing server sends.

The fourth aspect of the embodiment of the present invention, it is provided that a kind of text analyzing server, including:

Receiving unit, for receiving the text message that Call Center Server sends, described text message is that described Call Center Server is converted to according to voice messaging；

Text analysis unit, carries out text analyzing for the described text message that described reception unit is received, identifies described first participle set, comprise at least one text participle in described first participle set；

Transmitting element, for sending, to described Call Center Server, the described first participle set that described text analysis unit analysis obtains, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base.

In conjunction with fourth aspect, in the implementation that the first is possible, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Described text analyzing server, also includes:

Reading unit, for reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base；

Described text analysis unit, the described first retrieval data being additionally operable to described reading unit is read carry out text analyzing, identify the second participle set, comprise at least one text participle in described second participle set；

Generate unit, for the text participle in the described second participle set that obtains according to described text analysis unit analysis, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item, generates the described key word index list comprising at least one described key word index item；

Described transmitting element, is additionally operable to send, to described Call Center Server, the described key word index list that described generation unit generates.

Voice messaging is carried out the text message that text is converted to by data retrieval method and equipment, the voice messaging that Call Center Server reception user terminal sends, and acquisition that the embodiment of the present invention provides；Call Center Server obtains and text message is carried out the first participle set that text analyzing identifies, and comprises at least one text participle in first participle set；In the key word index list of the retrieval data that Call Center Server stores from knowledge data base, search the search key mated with the text participle in first participle set, key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with search key in key word index item, in knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；Call Center Server from described knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

nullWith time in prior art from knowledge data library searching data，It is too dependent on business experience and the technical ability of the individual of contact staff，The accuracy of retrieval result is relatively low with recall precision to be compared，Pass through this programme，Call Center Server is after receiving the voice messaging that user terminal sends，Can by voice messaging is carried out text analyzing obtain analyze result (first participle set)，Then from knowledge data base in the key word index list of the retrieval data of storage，Search the search key mated with the text participle in first participle set，Contact staff need not identify search key according to business experience and the technical ability of its people from user speech information，And what Call Center Server carried out search key automatically identifies the identification carrying out search key compared to contact staff's business experience according to its people and technical ability，The recognition efficiency of search key can be improved，And avoid the retrieval error owing to contact staff occurs due to carelessness or general idea，Improve the accuracy identifying search key，Efficiently、After being accurately obtained search key，Then can be quick、It is accurately obtained retrieval result.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

A kind of data retrieval method flow chart that Fig. 1 provides for the embodiment of the present invention one；

A kind of data retrieval method flow chart that Fig. 2 provides for the embodiment of the present invention two；

A kind of data retrieval method flow chart that Fig. 3 provides for the embodiment of the present invention three；

A kind of data retrieval method flow chart that Fig. 4 provides for the embodiment of the present invention four；

The structure composition schematic diagram of a kind of Call Center Server that Fig. 5 provides for the embodiment of the present invention five；

The structure composition schematic diagram of the another kind of Call Center Server that Fig. 6 provides for the embodiment of the present invention five；

The structure composition schematic diagram of the another kind of Call Center Server that Fig. 7 provides for the embodiment of the present invention five；

The structure composition schematic diagram of the another kind of Call Center Server that Fig. 8 provides for the embodiment of the present invention five；

The structure composition schematic diagram of a kind of text analyzing server that Fig. 9 provides for the embodiment of the present invention six；

The structure composition schematic diagram of the another kind of text analyzing server that Figure 10 provides for the embodiment of the present invention six；

The structure composition schematic diagram of a kind of Call Center Server that Figure 11 provides for the embodiment of the present invention seven；

The structure composition schematic diagram of the another kind of Call Center Server that Figure 12 provides for the embodiment of the present invention seven；

The structure composition schematic diagram of the another kind of Call Center Server that Figure 13 provides for the embodiment of the present invention seven；

The structure composition schematic diagram of a kind of text analyzing server that Figure 14 provides for the embodiment of the present invention eight；

The structure composition schematic diagram of a kind of data retrieval system that Figure 15 provides for the embodiment of the present invention eight.

Detailed description of the invention

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.

nullVarious technology described herein can be used for various wireless communication system，Such as current 2G，3G communication system and next generation communication system，Such as global system for mobile communications (GSM，GlobalSystemforMobilecommunications)，CDMA (CDMA，CodeDivisionMultipleAccess) system，Time division multiple acess (TDMA，TimeDivisionMultipleAccess) system，WCDMA (WCDMA，WidebandCodeDivisionMultipleAccessWireless)，Frequency division multiple access (FDMA，FrequencyDivisionMultipleAddressing) system，OFDM (OFDMA，OrthogonalFrequency-DivisionMultipleAccess) system，Single Carrier Frequency Division Multiple Access (SC-FDMA) system，GPRS (general packet radio service) (GPRS，GeneralPacketRadioService) system，Long Term Evolution (LTE，LongTermEvolution) system，And other these type of communication systems.

Herein in connection with terminal and/or base station and/or base station controller, various aspect is described.

User terminal, it is possible to be wireless terminal can also be catv terminal, wireless terminal may refer to provide a user with the equipment of voice and/or data connectivity, has the portable equipment of wireless connecting function or is connected to other process equipment of radio modem.Wireless terminal can through wireless access network (such as, RAN, RadioAccessNetwork) communicate with one or more core net, wireless terminal can be mobile terminal, such as mobile phone (or being called " honeycomb " phone) and the computer with mobile terminal, for example, it may be portable, pocket, hand-held, built-in computer or vehicle-mounted mobile device, they exchange language and/or data with wireless access network.Such as, PCS (PCS, PersonalCommunicationService) phone, wireless phone, Session initiation Protocol (SIP) phone, WLL (WLL, WirelessLocalLoop) stand, the equipment such as personal digital assistant (PDA, PersonalDigitalAssistant).Wireless terminal can also be called system, subscri er unit (SubscriberUnit), subscriber station (SubscriberStation), movement station (MobileStation), mobile platform (Mobile), distant station (RemoteStation), access point (AccessPoint), remote terminal (RemoteTerminal), access terminal (AccessTerminal), user terminal (UserTerminal), user agent (UserAgent), subscriber equipment (UserDevice), or subscriber's installation (UserEquipment).

It addition, the terms " system " and " network " are often used interchangeably in this article.The terms "and/or", is only a kind of incidence relation describing affiliated partner, and expression can exist three kinds of relations, for instance, A and/or B, it is possible to represent: individualism A, there is A and B, individualism B these three situation simultaneously.It addition, character "/" herein, typically represent forward-backward correlation to as if the relation of a kind of "or".

The method of the embodiment of the present invention is applied in call center system, call center system is to make full use of modern communication and computer technology, as interactive voice answering is (English: InteractiveVoiceResponse, it is called for short: IVR), automatically call distribution (English: AutomaticCallDistributor, ACD) it is called for short: the technology such as, it is possible to automatically and flexibly process the operation operating system of a large amount of various different phone incoming call and call business and service.

Wherein, existing call center system is mainly made up of Call Center Server and knowledge data base.User's query call is processed for the call center system of common carrier (providing the user the operator of communication service), contact staff after connected the voice communication with user by the Call Center Server in call center system, then can identify search key according to the business experience of individual and technical ability from user speech information；Then by search key Input knowledge data base, knowledge data base match the information mated with search key as retrieval data, and show retrieval data to contact staff；Contact staff also needs to retrieval extracting data that the business experience according to individual and technical ability shows from knowledge data base and the higher retrieval data of user speech information relevance, and make answer according to the retrieval data extracted for user, or by the retrieval data feedback that extracts to user.

Unlike the prior art, call center system in the embodiment of the present invention is when processing user's query call, then can realize voice messaging carrying out the text conversion of text conversion and text message carrying out information identification and the analytical work that text analyzing replaces contact staff to carry out in processing user's query script by Call Center Server, the accuracy of retrieval result can be improved, and improve recall precision.Now to data retrieval method provided by the invention with specific examples below and realize the equipment of this data retrieval method and illustrate:

Embodiment one

The embodiment of the present invention provides a kind of data retrieval method, as it is shown in figure 1, include:

S101, Call Center Server receive the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition.

Wherein, user can initiate voice communication request by user terminal to Call Center Server, after setting up the voice communication link between user terminal and Call Center Server, to the voice messaging that Call Center Server sends.Wherein, the voice messaging that user terminal sends is made up of at least one sound bite being dimension with sentence.

It should be noted that the Call Center Server in the embodiment of the present invention can be the operator that can provide the user communication service receives the call center management server that user calls.

S102, Call Center Server obtain and text message are carried out the first participle set that text analyzing identifies, and comprise at least one text participle in first participle set.

Wherein, Call Center Server obtain the method that text message carried out the first participle set that text analyzing identifies specifically may include that Call Center Server text message is made pauses in reading unpunctuated ancient writings, participle, to obtain at least one text participle.

Exemplary, text message can be carried out text analyzing by Call Center Server, identify the 3rd participle set (comprising at least one text participle in the 3rd participle set), then the meaningless text participle comprised in the 3rd participle set is filtered, obtain first participle set, it is not intended to Yi Wenben participle includes: exclamation and modal particle.

Exemplary, text message is carried out text analyzing by Call Center Server, identify the 3rd participle set (comprising at least one text participle in the 3rd participle set), then the first text participle comprised in the 3rd participle set is filtered, obtaining first participle set, the first text participle is text participle all unmatched with any one search key in key word index list in the 3rd participle set.

Further, in a kind of application scenarios of the embodiment of the present invention, Call Center Server can send text message to text analyzing server, text analyzing server text message is carried out text analyzing, identify first participle set；Then the first participle set that text analyzing server sends is received.

In the key word index list of the retrieval data that S103, Call Center Server store from knowledge data base, search the search key mated with the text participle in first participle set.

Wherein, comprising at least one key word index item in key word index list, comprise search key and the mark of the retrieval data corresponding with search key in key word index item, in knowledge data base, storage has at least one retrieval data and the mark of retrieval data.

In a kind of application scenarios of the embodiment of the present invention, key word index item can be that each retrieval data of storage in knowledge data base be carried out text analyzing by Call Center Server, the mark composition of the search key retrieval data corresponding with search key of each the retrieval data obtained.

Wherein, each retrieval data of storage in knowledge data base are carried out text analyzing by Call Center Server, the concrete grammar obtaining the search key of each retrieval data may include that Call Center Server reads the first retrieval data of storage in described knowledge data base, and described first retrieval data are any one retrieval data of storage in described knowledge data base；Obtain and described first retrieval data are carried out the second participle set that text analyzing identifies, described second participle set comprises at least one text participle；According to the text participle in described second participle set, generate the search key of described first retrieval data.

Further, in the another kind of application scenarios of the embodiment of the present invention, Call Center Server can receive the key word index list that text analyzing server sends.Wherein, the method that text analyzing server generates key word index list is similar with the method for the key word index list that Call Center Server generates, and the present embodiment repeats no more here.

S104, Call Center Server from knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Wherein, Call Center Server, then can from knowledge data base after finding the search key mated with the text participle in first participle set, the retrieval data indicated by mark of the data of the retrieval corresponding to search key that retrieval arrives.

The data retrieval method that the embodiment of the present invention provides, Call Center Server receives the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition；Call Center Server obtains and text message is carried out the first participle set that text analyzing identifies, and comprises at least one text participle in first participle set；In the key word index list of the retrieval data that Call Center Server stores from knowledge data base, search the search key mated with the text participle in first participle set, key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with search key in key word index item, in knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；Call Center Server from described knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Embodiment two

The embodiment of the present invention provides a kind of data retrieval method, as in figure 2 it is shown, include:

S201, text analyzing server receive the text message that Call Center Server sends, and text message is that Call Center Server is converted to according to voice messaging.

Wherein, text message is made up of at least one text fragments, and voice messaging is made up of at least one sound bite being dimension with sentence.

Call Center Server can in the process carrying out voice communication with user terminal, after receiving the sound bite that user terminal sends, then send sound bite to voice analysis server, so that sound bite is converted to text fragments by voice analysis server；Call Center Server after receiving the text fragments that voice analysis server sends, then can send text fragments to text analyzing server.

Text message is carried out text analyzing by S202, text analyzing server, identifies first participle set, comprises at least one text participle in first participle set.

Wherein, text analyzing server text message is carried out the method for the first participle set that text analyzing identifies specifically may include that text analyzing server text message is made pauses in reading unpunctuated ancient writings, participle, to identify at least one text participle.

Exemplary, text message can be carried out text analyzing by text analyzing server, identify the 3rd participle set (comprising at least one text participle in the 3rd participle set), then the meaningless text participle comprised in the 3rd participle set is filtered, obtain first participle set, it is not intended to Yi Wenben participle includes: exclamation and modal particle.

Exemplary, text message can be carried out text analyzing by text analyzing server, identify the 3rd participle set (comprising at least one text participle in the 3rd participle set), then the first text participle comprised in the 3rd participle set is filtered, obtaining first participle set, the first text participle is text participle all unmatched with any one search key in key word index list in the 3rd participle set.

S203, text analyzing server send first participle set to Call Center Server, so that Call Center Server retrieves data according to the text participle in first participle set from knowledge data base.

Further, each retrieval data of storage in knowledge data base can be carried out text analyzing by text analyzing server, obtain the search key of each retrieval data, then the key word index item of the mark of the search key comprising the first retrieval data and the first retrieval data is generated, and generate the key word index list comprising at least one key word index item, then key word index list is sent to Call Center Server, so that in the key word index list of retrieval data that can store from knowledge data base of Call Center Server, search the search key mated with the text participle in first participle set, and from knowledge data base, the retrieval data indicated by mark of the data of the retrieval corresponding to search key that retrieval arrives.

It is understood that the method for the present embodiment can also include: text analyzing server reads the first retrieval data of storage in knowledge data base, the first retrieval data are any one retrieval data of storage in knowledge data base；Text analyzing server is retrieved data to described first and is carried out text analyzing, identifies the second participle set, comprises at least one text participle in the second participle set；Text analyzing server, according to the text participle in the second participle set, generates the search key of the first retrieval data；Text analyzing server generates a key word index item according to the mark of the first search key retrieving data and the first retrieval data；Text server generates the key word index list comprising at least one key word index item；Text analyzing server sends key word index list to Call Center Server.

The data retrieval method that the embodiment of the present invention provides, text analyzing server receives the text message that Call Center Server sends, and described text message is that described Call Center Server is converted to according to voice messaging；Described text message is carried out text analyzing by described text analyzing server, identifies described first participle set, comprises at least one text participle in described first participle set；Described text analyzing server sends described first participle set to described Call Center Server, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base.

nullWith time in prior art from knowledge data library searching data，It is too dependent on business experience and the technical ability of the individual of contact staff，The accuracy of retrieval result is relatively low with recall precision to be compared，Pass through this programme，The text message that Call Center Server sends can be carried out text analyzing and obtain analyzing result (first participle set) by text analyzing server，And first participle set is sent to Call Center Server，In the key word index list of the retrieval data stored from knowledge data base by Call Center Server，Search the search key mated with the text participle in first participle set，Contact staff need not identify search key according to business experience and the technical ability of its people from user speech information，And what Call Center Server carried out search key automatically identifies the identification carrying out search key compared to contact staff's business experience according to its people and technical ability，The recognition efficiency of search key can be improved，And avoid the retrieval error owing to contact staff occurs due to carelessness or general idea，Improve the accuracy identifying search key，Efficiently、After being accurately obtained search key，Then can be quick、It is accurately obtained retrieval result.

Embodiment three

The embodiment of the present invention provides a kind of data retrieval method, it is applied to work as Call Center Server be integrated with the text conversion module being capable of voice messaging carries out text conversion and being capable of and voice messaging is carried out the text message that text is converted to by text conversion module carry out the text analysis model of text analyzing, namely Call Center Server can realize voice messaging is carried out text conversion, and then in the concrete scene of function carrying out the text message that text is converted to and carrying out text analyzing.

In this application scenarios, as it is shown on figure 3, this data retrieval method, it is possible to including:

S301, Call Center Server read the first retrieval data of storage in knowledge data base, and the first retrieval data are any one retrieval data of storage in knowledge data base.

S302, Call Center Server obtain and the first retrieval data are carried out text analyzing, and the second participle set identified comprises at least one text participle in the second participle set.

Wherein, any one retrieval data of storage in knowledge data base can be carried out text analyzing by Call Center Server, and the second participle set identified comprises at least one text participle in the second participle set.The method that the retrieval data of storage in knowledge data base are carried out text analyzing by Call Center Server is similar with the method that text message is carried out text analyzing by Call Center Server, the retrieval data of storage in knowledge data base are carried out the method for text analyzing and are referred in the embodiment of the present invention Call Center Server text message is carried out the concrete grammar of text analyzing by Call Center Server, and the embodiment of the present invention repeats no more here.

S303, Call Center Server, according to the text participle in the second participle set, generate the search key of the first retrieval data.

S304, Call Center Server generate a key word index item according to the mark of the first search key retrieving data and the first retrieval data.

S305, Call Center Server generate the key word index list comprising at least one key word index item.

Exemplary, the embodiment of the present invention, for table 1, gives the example of a key word index list:

Table 1:

S306, Call Center Server receive the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition.

Wherein, voice messaging can be carried out text conversion by Call Center Server, to obtain text message.Voice messaging can be carried out text and be converted to the method for text message and be referred in prior art to realize the concrete grammar of voice converting text by Call Center Server, and the embodiment of the present invention repeats no more here.

Exemplary, Call Center Server can be installed the application program possessing voice converting text function, to realize the voice messaging conversion to text message.

S307, Call Center Server obtain and text message are carried out the first participle set that text analyzing identifies, and comprise at least one text participle in first participle set.

Exemplary, voice messaging is carried out the text message that text is converted to getting by Call Center Server: after " trouble helps me to consult the rate situation of set meal A ", then can to text message: " trouble helps me to consult the rate situation of set meal A " makes pauses in reading unpunctuated ancient writings, participle, then can get by text participle: the first participle set that " looking into ", " set meal A " and " rate situation " is constituted.

Further, text message is being carried out after text analyzing obtains text participle by Call Center Server, it is also possible to filter out the meaningless text participle analyzed in the text participle obtained, to improve data search efficiency.Wherein, the meaningless text participle in the embodiment of the present invention may include that exclamation and modal particle；Meaningless text participle can also include greeting and term of courtesy etc..

Exemplary, assume that the text message that Call Center Server gets is after " you are good; me please be helped to consult the rate situation of set meal A; thanks ", then can to text message: " you are good; me please be helped to consult the rate situation of set meal A; thanks " carries out making pauses in reading unpunctuated ancient writings, participle, then can get by text participle: the 3rd participle set that " you are good " (greeting), " asking " (term of courtesy), " looking into ", " set meal A ", " rate situation ", " thanks " (term of courtesy) and " " (exclamation/modal particle) is constituted；Call Center Server can filter the greeting in the 3rd participle set: " you are good ", term of courtesy: " asking " and " thanks " and exclamation/modal particle: " ", obtains by text participle: the first participle set that " looking into ", " set meal A " and " rate situation " is constituted.

Further, Call Center Server after getting the 3rd participle set, can also search the search key mated with each the text participle in the 3rd participle set from key word index list；If any one search key in key word index list all divides word mismatch with a text in the 3rd participle set, Call Center Server then can filter text participle.

The method of the present embodiment can also include: text message is carried out text analyzing by Call Center Server, identify the 3rd participle set (comprising at least one text participle in the 3rd participle set), then the first text participle comprised in the 3rd participle set is filtered, obtaining first participle set, the first text participle is text participle all unmatched with any one search key in key word index list in the 3rd participle set.

Based on examples detailed above, assume any one search key in key word index list all with text participle: " looking into " does not mate, Call Center Server then can filter text participle: " looking into ", obtains by text participle: the first participle set that " set meal A " and " rate situation " is constituted.

It should be noted that S301-S305 can be first carried out in embodiments of the present invention, then perform S306-S307；S306-S307 can also be first carried out, then perform S301-S305.The embodiment of the present invention does not limit for S301-S305 and the S306-S307 sequencing performed.

In the key word index list of the retrieval data that S308, Call Center Server store from knowledge data base, search the search key mated with the text participle in first participle set.

S309, Call Center Server from knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Wherein, in knowledge data base, storage has at least one retrieval data and the mark of retrieval data.

Exemplary, the embodiment of the present invention is for table 2, and knowledge data base can adopt the mode of list to store multiple retrieval data and the mark of each retrieval data:

Table 2:

Embodiment four

The embodiment of the present invention provides a kind of data retrieval method, is applied to during Call Center Server do not possess and voice messaging carries out text conversion and text message carries out the concrete scene of function of text analyzing.

In this application scenarios, as shown in Figure 4, this data retrieval method, it is possible to including:

S401, text analyzing server read the first retrieval data of storage in knowledge data base, and the first retrieval data are any one retrieval data of storage in knowledge data base.

Exemplary, knowledge data base can adopt the mode of list to store multiple retrieval data and the mark of each retrieval data, and the concrete form of knowledge data base and content are referred to the tabular form example of the knowledge data base shown in embodiment of the present invention table 2.

S402, text analyzing server are retrieved data to first and are carried out text analyzing, identify the second participle set, comprise at least one text participle in the second participle set.

Wherein, the first retrieval data are any one retrieval data of storage in knowledge data base.Any one retrieval data of storage in knowledge data base can be carried out text analyzing by text analyzing server, and the second participle set identified comprises at least one text participle in the second participle set.The method that the retrieval data of storage in knowledge data base are carried out text analyzing by text analyzing server is similar with the method that text message is carried out text analyzing by Call Center Server, the retrieval data of storage in knowledge data base are carried out the method for text analyzing and are referred in the embodiment of the present invention Call Center Server text message is carried out the concrete grammar of text analyzing by text analyzing server, and the embodiment of the present invention repeats no more here.

S403, text analyzing server, according to the text participle in the second participle set, generate the search key of the first retrieval data.

S404, text analyzing server generate a key word index item according to the mark of the first search key retrieving data and the first retrieval data.

S405, text server generate the key word index list comprising at least one key word index item, and send, to Call Center Server, the key word index list that at least one key word index item is constituted.

The concrete form of key word index list and content are referred to the key word index list shown in embodiment of the present invention table 1.Further, text server generates the specific descriptions that the method for key word index list is referred in the embodiment of the present invention to Call Center Server generation key word index tabulating method, and the embodiment of the present invention repeats no more here.

S406, Call Center Server receive the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition.

Wherein, voice messaging is made up of at least one sound bite being dimension with sentence.

As shown in Figure 4, S406 specifically may include that S406a-S406d:

S406a, Call Center Server receive the sound bite that user terminal sends.

S406b, Call Center Server, after receiving a sound bite, send sound bite to voice analysis server.

Sound bite is converted to text fragments by S406c, voice analysis server.

Wherein, sound bite is converted to the method for text fragments and Call Center Server by voice analysis server, and that sound bite is converted to the method for text fragments is similar, and the embodiment of the present invention repeats no more here.

S406d, Call Center Server receive the text fragments that voice analysis server sends.

S407, Call Center Server send text message to text analyzing server.

Wherein, text message is made up of at least one text fragments, and Call Center Server after receiving the text fragments that voice analysis server sends, then can send text fragment to text analyzing server；Certainly, Call Center Server after receiving all text fragments constituting text message that voice analysis server sends, then can also send text information to text analyzing server.

Text message is carried out text analyzing by S408, text analyzing server, identifies first participle set.

It should be noted that, text message is carried out text analyzing by text analyzing server, identifying the method for first participle set to be referred to Call Center Server text message is carried out text analyzing, identify the concrete grammar of first participle set, the embodiment of the present invention repeats no more here.

S409, text analyzing server send first participle set to Call Center Server.

It should be noted that S401-S405 can be first carried out in embodiments of the present invention, then perform S406-S409；S406-S409 can also be first carried out, then perform S401-S405.The embodiment of the present invention does not limit for S401-S405 and the S406-S409 sequencing performed.

In the key word index list of the retrieval data that S410, Call Center Server store from knowledge data base, search the search key mated with the text participle in first participle set.

S411, Call Center Server from knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

The data retrieval method that the embodiment of the present invention provides, voice messaging can be sent to voice analysis server after receiving the voice messaging that user terminal sends by Call Center Server, by voice analysis server, voice messaging is carried out the text message that text is converted to, and via Call Center Server, text message is forwarded to text analyzing server, by text analyzing server, text message is carried out the first participle set that text analyzing identifies, first participle set comprises at least one text participle；In the key word index list of the retrieval data that Call Center Server stores from knowledge data base again, search the search key mated with the text participle in first participle set, key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；Call Center Server from knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Embodiment five

The embodiment of the present invention provides a kind of Call Center Server, in order to perform the data retrieval method shown in Fig. 1, as it is shown in figure 5, this Call Center Server includes: receive unit 51, text conversion units 52, text analysis unit 53, keyword search unit 54 and retrieval unit 55.

Receive unit 51, for receiving the voice messaging that user terminal sends.

Text conversion units 52, carries out, for obtaining the described voice messaging that described reception unit is received, the text message that text is converted to.

Text analysis unit 53, carries out, for obtaining the described text message that described text conversion units 52 is obtained, the first participle set that text analyzing identifies, comprises at least one text participle in described first participle set.

Keyword search unit 54, for in the key word index list of the retrieval data of storage from knowledge data base, search the search key that the text participle in the described first participle set obtained with described text analysis unit 53 mates, described key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data.

Retrieval unit 55, for, from described knowledge data base, retrieving the retrieval data that the mark of the data of the retrieval corresponding to search key that described keyword search unit 54 finds is indicated.

Wherein, the described voice messaging that described reception unit 51 receives is: user terminal after setting up the voice communication link between user terminal and Call Center Server, is received from user terminal after initiating voice communication request to Call Center Server.Wherein, the voice messaging that user terminal sends is made up of at least one sound bite being dimension with sentence.

Wherein, text message can be made pauses in reading unpunctuated ancient writings by text analysis unit 53, participle, to obtain at least one text participle.

Further, as shown in Figure 6, described Call Center Server, it is also possible to including: read unit 56 and generate unit 57.

Read unit 56, for in the key word index item in the key word index list of the retrieval data stored from described knowledge data base at described keyword search unit 54, before searching the search key mated with the text participle in described first participle set, reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base.

Described text analysis unit 53, is additionally operable to obtain the described first retrieval data that described reading unit 56 is read and carries out the second participle set that text analyzing identifies, comprise at least one text participle in described second participle set.

Generate unit 57, for the text participle in the described second participle set according to the acquisition of described text analysis unit 53, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item, generates the described key word index list comprising at least one described key word index item.

Further, described reception unit 51, it is additionally operable in the key word index item in the key word index list of retrieval data stored from described knowledge data base at described keyword search unit 54, before searching the search key mated with the text participle in described first participle set, receive the described key word index list that text analyzing server sends.

Further, in a kind of application scenarios of the embodiment of the present invention, as it is shown in fig. 7, described text analysis unit 53, including: text analysis model 531 and the first filtering module 532.

Text analysis model 531, for described text message is carried out text analyzing, identifies the 3rd participle set, comprises at least one text participle in described 3rd participle set.

First filtering module 532, for filtering the meaningless text participle comprised in the described 3rd participle set that described text analysis model 531 identifies, obtains described first participle set, and described meaningless text participle includes: exclamation and modal particle.

Further, in the another kind of application scenarios of the embodiment of the present invention, as shown in Figure 8, described text analysis unit, including: text analysis model 531 and the second filtering module 533.

Second filtering module 533, the the first text participle comprised in the 531 described 3rd participle set not gone out for filtering described text analysis model to know, obtaining described first participle set, described first text participle is text participle all unmatched with any one search key in described key word index list in described 3rd participle set.

Further, in the another kind of application scenarios of the embodiment of the present invention, described text analysis unit, including: text analysis model the 531, first filtering module 532 and the second filtering module 533.

In this application scenarios, while the first filtering module 532 filters the meaningless text participle comprised in the described 3rd participle set that described text analysis model 531 identifies, the second filtering module 533 filters described text analysis model and knows the first text participle comprised in the 531 described 3rd participle set not gone out；Now, the participle intersection of sets collection that the participle set that the first filtering module 532 is filtrated to get and the second filtering module 533 are filtrated to get then can as described first participle set.

Further, described voice messaging is made up of at least one sound bite being dimension with sentence, and described text message is made up of at least one text fragments.

Described reception unit 51, specifically for receiving the described sound bite that described user terminal sends.

Described text conversion units 52, specifically for, after described reception unit 51 receives a described sound bite, sending described sound bite to voice analysis server, so that described sound bite is converted to described text fragments by described voice analysis server；Receive the described text fragments that described voice analysis server sends.

Further, described text analysis unit 53, specifically for sending described text message to described text analyzing server, so that described text message is carried out text analyzing by described text analyzing server, identifies described first participle set；Receive the described first participle set that described text analyzing server sends.

It should be noted that the Call Center Server in the embodiment of the present invention can be the operator that can provide the user communication service receives the call center management server that user calls.In the Call Center Server that the embodiment of the present invention provides, the specific descriptions of part functional module are referred to the corresponding content in embodiment of the method, and the present embodiment is no longer described in detail here.

The Call Center Server that the embodiment of the present invention provides, Call Center Server receives the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition；Call Center Server obtains and text message is carried out the first participle set that text analyzing identifies, and comprises at least one text participle in first participle set；In the key word index list of the retrieval data that Call Center Server stores from knowledge data base, search the search key mated with the text participle in first participle set, key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with search key in key word index item, in knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；Call Center Server from described knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Embodiment six

The embodiment of the present invention provides a kind of text analyzing server, in order to perform the data retrieval method shown in Fig. 2, as it is shown in figure 9, text Analysis server includes: receive unit 61, text analysis unit 62 and transmitting element 63.

Receiving unit 61, for receiving the text message that Call Center Server sends, described text message is that described Call Center Server is converted to according to voice messaging.

Text analysis unit 62, carries out text analyzing for the described text message that described reception unit 61 is received, identifies described first participle set, comprise at least one text participle in described first participle set.

Transmitting element 63, the described first participle set obtained is analyzed, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base for sending described text analysis unit 62 to described Call Center Server.

Wherein, text message can be made pauses in reading unpunctuated ancient writings by text analysis unit 62, participle, to identify at least one text participle.

Further, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data.

As shown in Figure 10, described text analyzing server, it is also possible to including: read unit 64 and generate unit 65.

Reading unit 64, for reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base.

Described text analysis unit 63, the described first retrieval data being additionally operable to described reading unit is read carry out text analyzing, identify the second participle set, comprise at least one text participle in described second participle set.

Generate unit 65, for analyzing the text participle in the described second participle set obtained according to described text analysis unit 63, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item, generates the described key word index list comprising at least one described key word index item.

Described transmitting element 63, is additionally operable to send, to described Call Center Server, the described key word index list that described generation unit 65 generates.

It should be noted that the specific descriptions of part functional module are referred to the corresponding content in embodiment of the method in the text analyzing server of embodiment of the present invention offer, the present embodiment is no longer described in detail here.

The text analyzing server that the embodiment of the present invention provides, it is possible to receiving the text message that Call Center Server sends, described text message is that described Call Center Server is converted to according to voice messaging；Described text message is carried out text analyzing by described text analyzing server, identifies described first participle set, comprises at least one text participle in described first participle set；Described text analyzing server sends described first participle set to described Call Center Server, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base.

Embodiment seven

On hardware realizes, unit in embodiment five can be embedded in or in the form of hardware independent of in the processor of Call Center Server, can also be stored in the memorizer of Call Center Server in a software form, so that processor calls performs the operation that above unit is corresponding, this processor can be CPU (CPU), microprocessor, single-chip microcomputer etc..

The embodiment of the present invention provides a kind of Call Center Server, and as shown in figure 11, this Call Center Server includes: receptor 71 and processor 72.

Receptor 71, for receiving the voice messaging that user terminal sends.

Processor 72, carries out, for obtaining the described voice messaging that described receptor 71 is received, the text message that text is converted to；Obtain and described text message is carried out the first participle set that text analyzing identifies, described first participle set comprises at least one text participle；From knowledge data base in the key word index list of the retrieval data of storage, search the search key mated with the text participle in described first participle set, described key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；From described knowledge data base, the retrieval data indicated by mark of the data of the retrieval corresponding to search key that retrieval arrives.

Further, in a kind of application scenarios of the embodiment of the present invention, described processor 72, it is additionally operable in the key word index item in the key word index list of the retrieval data of storage from knowledge data base, before searching the search key mated with the text participle in described first participle set, reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base；Obtain and described first retrieval data are carried out the second participle set that text analyzing identifies, described second participle set comprises at least one text participle；According to the text participle in described second participle set, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item；Generate the described key word index list comprising at least one described key word index item.

In the another kind of application scenarios of the embodiment of the present invention, described receptor 71, it is additionally operable in the key word index item in the key word index list of retrieval data stored from described knowledge data base at described processor 72, before searching the search key mated with the text participle in described first participle set, receive the described key word index list that text analyzing server sends.

In above two application scenarios, as shown in figure 12, described Call Center Server, it is also possible to including: memorizer 73.

Memorizer 73, for storing the described key word index list that described processor 72 generates；Or, memorizer 73, for storing the described key word index list that described receptor 71 receives.

Further, described processor 72, specifically for described text message is carried out text analyzing, identifies the 3rd participle set, comprises at least one text participle in described 3rd participle set；Filtering the meaningless text participle comprised in described 3rd participle set, obtain described first participle set, described meaningless text participle includes: exclamation and modal particle.

Further, described processor 72, specifically for described text message is carried out text analyzing, identifies the 3rd participle set, comprises at least one text participle in described 3rd participle set；Filtering the first text participle comprised in described 3rd participle set, obtain described first participle set, described first text participle is text participle all unmatched with any one search key in described key word index list in described 3rd participle set.

Described receptor 71, is additionally operable to receive the described sound bite that described user terminal sends.

As shown in figure 13, described Call Center Server, it is also possible to including: transmitter 74.

Transmitter 74, for, after described receptor 71 receives a described sound bite, sending described sound bite to voice analysis server, so that described sound bite is converted to described text fragments by described voice analysis server.

Described receptor 71, is additionally operable to receive the described text fragments that described voice analysis server sends.

Further, described transmitter 74, it is additionally operable to send described text message to described text analyzing server, so that described text message is carried out text analyzing by described text analyzing server, identifies described first participle set.

Described receptor 71, is additionally operable to receive the described first participle set that described text analyzing server sends.

In embodiments of the present invention, receptor 71, processor 72, memorizer 73 and transmitter 74 are connected by bus and complete mutual communicating.

Wherein, bus can be industry standard architecture (IndustryStandardArchitecture, ISA) total, external equipment interconnection (PeripheralComponentInterconnect, PCI) bus or extended industry-standard architecture (ExtendedIndustryStandardArchitecture, EISA) bus etc..This bus can be divided into address bus, data/address bus, control bus etc..For ease of representing, Figure 11-Figure 13 only represents with a thick line, it is not intended that only have a bus or a type of bus.

Processor 72 can be central processing unit (CentralProcessingUnit, CPU), it can also be the combination of control chip (such as baseband chip) in CPU, digital signal processor (DigitalSignalProcessor, DSP) and communication unit.In embodiments of the present invention, CPU can be single arithmetic core, it is also possible to include multioperation core.

Receptor 71 and transmitter 74 can be the device being sent and receiving wireless signal by antenna, it is also possible to provide signal to send the device with receiving interface for other.

It should be noted that the specific descriptions of part functional module are referred to the corresponding content in embodiment of the method in the Call Center Server of embodiment of the present invention offer, the present embodiment is no longer described in detail here.

The Call Center Server that the embodiment of the present invention provides, it is possible to receive the voice messaging that user terminal sends, and voice messaging is carried out the text message that text is converted to by acquisition；Call Center Server obtains and text message is carried out the first participle set that text analyzing identifies, and comprises at least one text participle in first participle set；In the key word index list of the retrieval data that Call Center Server stores from knowledge data base, search the search key mated with the text participle in first participle set, key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with search key in key word index item, in knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；Call Center Server from described knowledge data base, the retrieval data that the mark of the data of the retrieval corresponding to search key that retrieval arrives is indicated.

Embodiment eight

On hardware realizes, unit in embodiment six can be embedded in or in processor independent of text analyzing server in the form of hardware, can also be stored in a software form in the memorizer of text analyzing server, so that processor calls performs the operation that above unit is corresponding, this processor can be CPU (CPU), microprocessor, single-chip microcomputer etc..

The embodiment of the present invention provides a kind of text analyzing server, as shown in figure 14, including: receptor 81, processor 82 and transmitter 83.

Receptor 81, for receiving the text message that Call Center Server sends, described text message is that described Call Center Server is converted to according to voice messaging.

Processor 82, carries out text analyzing for the described text message that described receptor 81 is received, identifies described first participle set, comprise at least one text participle in described first participle set.

Transmitter 83, for sending, to described Call Center Server, the described first participle set that described processor 82 identifies, so that described Call Center Server retrieves data according to the text participle in described first participle set from knowledge data base.

Described processor 82, is additionally operable to read the first retrieval data of storage in described knowledge data base, and described first retrieval data are any one retrieval data of storage in described knowledge data base；Retrieve data to described first and carry out text analyzing, identify the second participle set, described second participle set comprises at least one text participle；According to the text participle in described second participle set, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item；Generate the described key word index list comprising at least one described key word index item.

Described transmitter 83, is additionally operable to send, to described Call Center Server, the described key word index list that described processor 82 generates.

In embodiments of the present invention, receptor 81, processor 82, transmitter 83 are connected by bus and complete mutual communication.

Wherein, bus can be industry standard architecture (IndustryStandardArchitecture, ISA) total, external equipment interconnection (PeripheralComponentInterconnect, PCI) bus or extended industry-standard architecture (ExtendedIndustryStandardArchitecture, EISA) bus etc..This bus can be divided into address bus, data/address bus, control bus etc..For ease of representing, Figure 14 only represents with a thick line, it is not intended that only have a bus or a type of bus.

Processor 82 can be central processing unit (CentralProcessingUnit, CPU), it can also be the combination of control chip (such as baseband chip) in CPU, digital signal processor (DigitalSignalProcessor, DSP) and communication unit.In embodiments of the present invention, CPU can be single arithmetic core, it is also possible to include multioperation core.

Receptor 81 and transmitter 82 can be the device being sent and receiving wireless signal by antenna, it is also possible to provide signal to send the device with receiving interface for other.

It should be noted that, in the first application scenarios of the embodiment of the present invention, Call Center Server can be integrated with the text conversion module being capable of voice messaging carries out text conversion and being capable of voice messaging is carried out the text message that text is converted to by text conversion module carry out the text analysis model of text analyzing, namely can realize voice messaging is carried out text conversion inside Call Center Server, and then to carrying out the text message that text is converted to and carry out the function of text analyzing.

In the second application scenarios of the embodiment of the present invention, Call Center Server needs to send voice messaging to voice analysis server, realized voice messaging is carried out text conversion by voice analysis server, obtain text message, then receive the text message that voice analysis server sends；Call Center Server needs to send text message to text analyzing server, text analyzing server realize text message is carried out text analyzing, obtains analyzing result, then receives the analysis result that text analyzing server sends.

Corresponding to the second application scenarios, the embodiment of the present invention provides a kind of data retrieval system, and as shown in figure 15, this data retrieval system includes: Call Center Server 91, voice analysis server 92, text analyzing server 93 and knowledge data base 94.

Described Call Center Server 91, for receiving the voice messaging that user terminal sends, and sends described voice messaging to described voice analysis server 92, and described voice messaging is made up of at least one sound bite being dimension with sentence.

Described voice analysis server 92, for described sound bite is converted to described text fragments, and sends text message to described Call Center Server 91, and described text message is made up of at least one text fragments.

Described Call Center Server 91, is additionally operable to send described text message to described text analyzing server 93.

Described text analyzing server 93, for described text message is carried out text analyzing, identifies first participle set, comprises at least one text participle in described first participle set.

Described Call Center Server 91, it is additionally operable to from knowledge data base in the key word index list of the retrieval data of storage, search the search key mated with the text participle in described first participle set, described key word index list comprises at least one key word index item, comprising search key and the mark of the retrieval data corresponding with described search key in described key word index item, in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；From described knowledge data base, the retrieval data indicated by mark of the data of the retrieval corresponding to search key that retrieval arrives.

Further, described text analyzing server 93, it is additionally operable to send described key word index list to described Call Center Server 91.

Further, described text analyzing server 93, it is additionally operable to before sending described key word index list to described Call Center Server 91, reading the first retrieval data of storage in described knowledge data base, described first retrieval data are any one retrieval data of storage in described knowledge data base；Retrieve data to described first and carry out text analyzing, identify the second participle set, described second participle set comprises at least one text participle；According to the text participle in described second participle set, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item；Generate the described key word index list comprising at least one described key word index item.

nullThe data retrieval system that the embodiment of the present invention provides，With time in prior art from knowledge data library searching data，It is too dependent on business experience and the technical ability of the individual of contact staff，The accuracy of retrieval result is relatively low with recall precision to be compared，Pass through this programme，Call Center Server is after receiving the voice messaging that user terminal sends，Can by voice messaging is carried out text analyzing obtain analyze result (first participle set)，Then from knowledge data base in the key word index list of the retrieval data of storage，Search the search key mated with the text participle in first participle set，Contact staff need not identify search key according to business experience and the technical ability of its people from user speech information，And what Call Center Server carried out search key automatically identifies the identification carrying out search key compared to contact staff's business experience according to its people and technical ability，The recognition efficiency of search key can be improved，And avoid the retrieval error owing to contact staff occurs due to carelessness or general idea，Improve the accuracy identifying search key，Efficiently、After being accurately obtained search key，Then can be quick、It is accurately obtained retrieval result.

Through the above description of the embodiments, those skilled in the art is it can be understood that arrive, for convenience and simplicity of description, only it is illustrated with the division of above-mentioned each functional module, in practical application, as desired above-mentioned functions distribution can be completed by different functional modules, be divided into different functional modules by the internal structure of device, to complete all or part of function described above.The specific works process of the system of foregoing description, device and unit, it is possible to reference to the corresponding process in preceding method embodiment, do not repeat them here.

In several embodiments provided herein, it should be understood that disclosed system, apparatus and method, it is possible to realize by another way.Such as, device embodiment described above is merely schematic, such as, the division of described module or unit, being only a kind of logic function to divide, actual can have other dividing mode when realizing, for instance multiple unit or assembly can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some interfaces, device or unit, it is possible to be electrical, machinery or other form.

The described unit illustrated as separating component can be or may not be physically separate, and the parts shown as unit can be or may not be physical location, namely may be located at a place, or can also be distributed on multiple NE.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.

It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to be that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated unit both can adopt the form of hardware to realize, it would however also be possible to employ the form of SFU software functional unit realizes.

If described integrated unit is using the form realization of SFU software functional unit and as independent production marketing or use, it is possible to be stored in a computer read/write memory medium.Based on such understanding, part or all or part of of this technical scheme that prior art is contributed by technical scheme substantially in other words can embody with the form of software product, this computer software product is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) or processor (processor) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-OnlyMemory), the various media that can store program code such as random access memory (RAM, RandomAccessMemory), magnetic disc or CD.

The above; being only the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, any those familiar with the art is in the technical scope that the invention discloses; change can be readily occurred in or replace, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.

Claims

1. a data retrieval method, it is characterised in that including:

2. method according to claim 1, it is characterized in that, in key word index item in the key word index list of the retrieval data stored from knowledge data base at described Call Center Server, before searching the search key mated with the text participle in described first participle set, described method also includes:

Described Call Center Server generates a described key word index item according to the mark of the described first search key retrieving data and described first retrieval data；

Described Call Center Server generates the described key word index list comprising at least one described key word index item.

3. method according to claim 1, it is characterized in that, in key word index item in the key word index list of the retrieval data stored from described knowledge data base at described Call Center Server, before searching the search key mated with the text participle in described first participle set, described method also includes:

Described Call Center Server receives the described key word index list that text analyzing server sends.

4. the method according to any one of claim 1-3, it is characterised in that described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

5. the method according to any one of claim 1-3, it is characterised in that described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

6. the method according to claim 1 or 3, it is characterised in that described voice messaging is made up of at least one sound bite being dimension with sentence, and described text message is made up of at least one text fragments；

7. the method according to claim 1 or 3, it is characterised in that described Call Center Server obtains and described text message is carried out the first participle set that text analyzing identifies, including:

8. a data retrieval method, it is characterised in that including:

9. method according to claim 8, it is characterised in that in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Described method also includes:

Described text analyzing server generates a described key word index item according to the mark of the described first search key retrieving data and described first retrieval data；

Described text server generates the described key word index list comprising at least one described key word index item；

10. a Call Center Server, it is characterised in that including:

Receive unit, for receiving the voice messaging that user terminal sends；

11. Call Center Server according to claim 10, it is characterised in that also include:

Generate unit, for the text participle in the described second participle set that obtains according to described text analysis unit, generate the search key of described first retrieval data；Mark according to the described first search key retrieving data and described first retrieval data generates a described key word index item, generates the described key word index list comprising at least one described key word index item.

12. Call Center Server according to claim 10, it is characterized in that, described reception unit, it is additionally operable in the key word index item in the key word index list of retrieval data stored from described knowledge data base at described keyword search unit, before searching the search key mated with the text participle in described first participle set, receive the described key word index list that text analyzing server sends.

13. the Call Center Server according to any one of claim 10-12, it is characterised in that described text analysis unit, including:

14. the Call Center Server according to any one of claim 10-12, it is characterised in that described text analysis unit, including:

15. the Call Center Server according to claim 10 or 12, it is characterised in that described voice messaging is made up of at least one sound bite being dimension with sentence, and described text message is made up of at least one text fragments；

16. the Call Center Server according to claim 10 or 12, it is characterized in that, described text analysis unit, specifically for sending described text message to described text analyzing server, so that described text message is carried out text analyzing by described text analyzing server, identify described first participle set；Receive the described first participle set that described text analyzing server sends.

17. a text analyzing server, it is characterised in that including:

18. text analyzing server according to claim 17, it is characterised in that in described knowledge data base, storage has the mark of at least one retrieval data and described retrieval data；

Described text analyzing server, also includes: