CN108984777B - Customer service method, apparatus and computer-readable storage medium - Google Patents
Customer service method, apparatus and computer-readable storage medium Download PDFInfo
- Publication number
- CN108984777B CN108984777B CN201810824060.3A CN201810824060A CN108984777B CN 108984777 B CN108984777 B CN 108984777B CN 201810824060 A CN201810824060 A CN 201810824060A CN 108984777 B CN108984777 B CN 108984777B
- Authority
- CN
- China
- Prior art keywords
- user
- characters
- key
- participles
- commodity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Character Discrimination (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a customer service method, a customer service device and a computer readable storage medium, and relates to the technical field of computers. The method of the invention comprises the following steps: identifying characters in a picture sent by a user in a customer service process; matching the characters with an object database, and identifying objects corresponding to the characters, wherein the object database comprises object information in an electronic commerce platform; and providing the description information of the object to the user so that the user can confirm the object to be consulted as the object. The method of the invention can intelligently identify the user intention, determine the object to be consulted by the user and make a corresponding response under the condition that the user sends the picture. The method of the invention enables the user to send the picture instead of manual input, thereby facilitating the consultation of the user, improving the service quality of the client and improving the user experience.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a method and an apparatus for customer service, and a computer-readable storage medium.
Background
With the rapid development of artificial intelligence technology, the business of customer service robots is developing at a high speed. Improving the quality of service of customer services has become an important goal in the management of e-commerce platforms.
At present, the customer service robot recognizes an object that a user wants to consult by recognizing characters input by the user to understand the intention of the user.
Disclosure of Invention
The inventor finds that: in the practical application process, a user can send pictures such as a commodity screenshot or an order screenshot to a client service robot for consultation conveniently, and the current client service robot cannot make a user-expected reaction on the pictures, so that the user experience is poor.
The invention aims to solve the technical problems that: how to improve the quality of service of the customer and improve the user experience.
According to some embodiments of the invention, there is provided a customer service method comprising: identifying characters in a picture sent by a user in a customer service process; matching the characters with an object database, and identifying objects corresponding to the characters, wherein the object database comprises object information in an electronic commerce platform; and providing the description information of the object to the user so that the user can confirm the object to be consulted as the object.
In some embodiments, matching the character to a database of objects, identifying the object to which the character corresponds comprises: performing word segmentation on the characters, filtering stop words and correcting wrong characters to obtain key word segments; and matching the key word segmentation with the object in the object database, and identifying the object corresponding to the key word segmentation.
In some embodiments, the object database comprises at least one of a name of a good in the e-commerce platform, a name of a good in the user order, and a name of a good in the user service order; matching the key participles with objects in an object database, and identifying the objects corresponding to the key participles comprises: under the condition that the key word is a character, matching the key word with the commodity name in the object database, and identifying at least one item of the commodity, the order and the service list corresponding to the key word; or the object database comprises commodity numbers, service order numbers or order numbers in the electronic commerce platform; matching the key participles with objects in an object database, and identifying the objects corresponding to the key participles comprises: and under the condition that the key word is a number, matching the key word with a commodity number, a service order number or an order number in the object database according to a numbering rule, and determining at least one of a commodity, an order and a service order corresponding to the key word.
In some embodiments, correcting the erroneous character includes: inputting the participles after the stop words are filtered into a pre-trained error correction model to obtain the correct probability of each participle, taking the participles with the correct probability lower than a threshold value as error participles, replacing the error participles with words with similar shapes, re-inputting the error correction model and determining the correct probability of each participle until the participles with the highest correct probability are selected as key participles; or matching the participles after the stop words are filtered with a wrong word bank, determining wrong participles and correcting the wrong participles.
In some embodiments, providing the description information of the object to the user comprises: matching the object with an object in a user consultation context; providing the description information of the object to the user in the case of matching to the same object; and under the condition that the same object is not matched, inquiring the user whether the object to be consulted is the object or not, and under the condition that the user confirms that the object to be consulted is the object, providing the description information of the object to the user.
In some embodiments, providing the description information of the object to the user comprises: providing each object to the user in a list form when the number of the objects is multiple; and acquiring the object to be consulted selected by the user from the list, and providing the description information of the object to be consulted to the user.
In some embodiments, the text in the picture is recognized according to an optical character recognition algorithm; or the description information of the object comprises at least one item of information of the name, the picture, the corresponding commodity number, the corresponding order number and the corresponding service order number of the object.
According to further embodiments of the present invention, there is provided a customer service device including: the character recognition module is used for recognizing characters in pictures sent by a user in the customer service process; the object identification module is used for matching the characters with an object database and identifying the objects corresponding to the characters, wherein the object database comprises object information in the e-commerce platform; and the information sending module is used for providing the description information of the object to the user so that the user can confirm that the object to be consulted is the object.
In some embodiments, the object recognition module is configured to perform word segmentation on the characters, filter stop words, and correct wrong characters to obtain key word segments; and matching the key word segmentation with the object in the object database, and identifying the object corresponding to the key word segmentation.
In some embodiments, the object database comprises at least one of a name of a good in the e-commerce platform, a name of a good in the user order, and a name of a good in the user service order; the object identification module is used for matching the key word with the commodity name in the object database under the condition that the key word is a character, and identifying at least one item of the commodity, the order and the service order corresponding to the key word; or the object database comprises commodity numbers, service order numbers or order numbers in the electronic commerce platform; and the object identification module is used for matching the key word with the commodity number, the service order number or the order number in the object database according to the numbering rule under the condition that the key word is the number, and determining at least one item of the commodity, the order and the service order corresponding to the key word.
In some embodiments, the object recognition module is configured to input the segmented words after the stop words are filtered into a pre-trained error correction model to obtain a correct probability of each segmented word, use the segmented words with the correct probability lower than a threshold as error segmented words, replace the error segmented words with similar shapes, re-input the error correction model and determine the correct probability of each segmented word until the segmented word with the highest correct probability is selected as a key segmented word; or the object recognition module is used for matching the segmented words after the stop words are filtered with the wrong word bank, determining the wrong segmented words and correcting the wrong segmented words.
In some embodiments, the information sending module is to match the object with an object in the user consultation context; providing the description information of the object to the user in the case of matching to the same object; and under the condition that the same object is not matched, inquiring the user whether the object to be consulted is the object or not, and under the condition that the user confirms that the object to be consulted is the object, providing the description information of the object to the user.
In some embodiments, the information sending module is configured to provide each object to the user in a list form if the object is multiple; and acquiring the object to be consulted selected by the user from the list, and providing the description information of the object to be consulted to the user.
In some embodiments, the text in the picture is recognized according to an optical character recognition algorithm; or the description information of the object comprises at least one item of information of the name, the picture, the corresponding commodity number, the corresponding order number and the corresponding service order number of the object.
According to still further embodiments of the present invention, there is provided a customer service device including: a memory; and a processor coupled to the memory, the processor configured to execute a customer service method as implementing any of the foregoing embodiments based on instructions stored in the memory device.
According to still further embodiments of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the customer service method of any of the preceding embodiments.
In the invention, after receiving the picture sent by the user in the customer service process, the characters in the picture are identified, the object corresponding to the characters is identified through the matching of the characters and the object database, and further the description information of the object can be provided for the user, so that the user can confirm that the object to be consulted is the object. The method of the invention can intelligently identify the user intention, determine the object to be consulted by the user and make a corresponding response under the condition that the user sends the picture. The method of the invention enables the user to send the picture instead of manual input, thereby facilitating the consultation of the user, improving the service quality of the client and improving the user experience.
Other features of the present invention and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 illustrates a flow diagram of a customer service method of some embodiments of the invention.
FIG. 2 shows a flow diagram of a customer service method of further embodiments of the present invention.
FIG. 3 illustrates a schematic diagram of a customer service device according to some embodiments of the invention.
FIG. 4 shows a schematic block diagram of a customer service device according to further embodiments of the present invention.
Fig. 5 is a schematic structural diagram of a customer service device according to further embodiments of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a customer service method, which can be realized by a customer service device (customer service robot), belonging to an automatic customer service method, wherein the automatic customer service can also be called automatic customer service, self-service customer service, electronic customer service, intelligent customer service, machine customer service, online customer service or automatic response customer service, and the like, and the following describes some embodiments with reference to fig. 1.
FIG. 1 is a flow chart of some embodiments of a customer service method of the present invention. As shown in fig. 1, the method of this embodiment includes: steps S102 to S106.
In step S102, characters in a picture sent by a user during customer service are identified.
When a user browses, purchases and the like on the e-commerce platform, the user can consult goods, orders or service lists (such as pre-sale or post-sale consultation service lists and the like). The user can select non-manual customer service for consultation, and pictures such as commodity screenshots, order screenshots or service order screenshots and the like can be sent to the customer service device in the service process. After receiving the picture sent by the user, the electronic customer service device may identify characters in the picture by using an OCR (Optical Character Recognition) algorithm, and convert pixels in the picture into characters represented by data that can be identified by a computer.
OCR algorithms typically include image preprocessing (e.g., binarization, denoising, tilt correction, etc.), layout segmentation, character feature extraction, character recognition (e.g., recognizing characters using a support vector machine or a neural network model), post-processing, and the like. Through the process, the picture expressed by pixels can be converted into editable characters, and the characters can be expressed by specific binary codes inside the computer and displayed as editable texts. The OCR algorithm may refer to an existing implementation and will not be described herein.
In step S104, the character is matched with the object database, and the object corresponding to the character is identified.
The object database comprises object information in the e-commerce platform, wherein the object is at least one of a commodity, an order and a service order. The object information may be at least one item of information of a commodity name, a commodity number, an order number, and a service order number. Further, the object information may be a product title or page layout information. The user can consult with any object of the e-commerce platform, and the information capable of reflecting the characteristics of the object can be stored in the object database, which is not limited to the illustrated example.
The characters after OCR recognition can be subjected to processes of word segmentation, stop word filtering, error character correction, repeated word removal and the like to obtain key word segmentation, and then the key word segmentation is matched with the object database, so that matching work can be reduced, and related objects can be matched more efficiently and accurately. The word segmentation can adopt the existing algorithm, and is not described in detail herein. Stop words such as special symbols, fictional words, etc. have little effect on matching objects. A dead word bank may be set, and dead words may be identified and filtered by comparing the segmented words with the dead word bank.
Characters identified by the OCR may have wrong characters, and the situation that objects cannot be matched or are matched wrongly easily occurs when object matching is carried out subsequently. Therefore, the method can carry out the processing of correcting wrong characters on the participles after the participles and the stop words are filtered.
In some embodiments, the erroneous characters may be corrected using a pre-trained error correction model. For example, the following method can be used to correct the erroneous characters in the segmented word after the stop word is filtered. (1) And inputting the participles into an error correction model to obtain the correct probability of each participle. (2) And selecting the participles with the correct probability lower than a threshold value as error participles. (3) And (4) judging whether a word capable of replacing the error word is existed or not, if so, executing the step (4), otherwise, executing the step (5). (4) And (4) replacing the wrong participle by using the similar-shaped word, and returning to the step (1) to restart the execution. (5) And selecting the participle under the condition of the highest correct probability of each participle as a key participle.
Further, the error correction model may be a combination of word2vector and LR (logistic regression) model. Firstly, a vector of word segmentation can be generated through a trained word2vector, and the vector can reflect the relevance of words before and after word segmentation. The vector of the word segmentation is input into the well-trained LR model, and the correct probability of the word segmentation can be obtained. The LR model is trained using manually labeled training samples, which may include various user consulted sentences and participles, where each participle is labeled as correct or incorrect. The neural network model can also be used as an error correction model to predict the correct probability for the vector of the participle, and is not limited to the illustrated example. For the wrong word segmentation, words with similar shapes can be used for replacement, and the corresponding relation of each shape word or shape word can be stored in advance.
For example, a word of 'rice' is recognized as 'Taimi', the word can be divided into two words of 'Tai' and 'Mi' during word segmentation, the error correction model is used for inputting the 'Tai' into the model, because the word2vector can reflect the relevance of words before and after word segmentation, the pre-trained model can recognize that the correct probability of the 'Tai' is lower, namely the smaller probability of the word appears in the sentence, therefore, the 'Tai' can be replaced by a 'big' shape similar to the word, and the error correction model is input again to obtain the higher correct probability. And after the shape word or the shape word replaces the error word segmentation, the word segmentation processing can be carried out again, and then the error correction model is input to calculate the correct probability of the word segmentation.
In other embodiments, a wrong word bank may be provided, which contains common wrong words and corresponding correct words, and the wrong characters are corrected by using the wrong word bank. For example, the segmented words after the stop words are filtered are matched with the wrong word bank, and the wrong segmented words are determined and corrected. When the segmentation is matched with the wrong word bank, other characters in a preset range before and after the segmentation can participate in matching with the segmentation, because the wrong words can be divided into different segmentation during the segmentation.
In still other embodiments, a sensitive word in the segmented word may be identified, and whether the segmented word is a sensitive word may be determined by comparing the segmented word with words in a sensitive word lexicon. In the case where the characters contain sensitive words, the user may be prompted to manually enter information for consultation.
By the method in the embodiment, the characters identified by the OCR are segmented, the stop words are filtered, the error characters are corrected, repeated segmentation can be removed, and the key segmentation is obtained. And matching the key word segmentation with the object database to determine a corresponding object. Some embodiments of matching key tokens to object databases are described below.
In some embodiments, the object database includes at least one of a name of a good in the e-commerce platform, a name of a good in the user order, and a name of a good in the user service order. And under the condition that the key word is a character, matching the key word with the commodity name in the object database, and identifying at least one item of the commodity, the order and the service list corresponding to the key word.
The key word segmentation can be matched with the commodity name in the electronic commerce platform, and meanwhile, the key word segmentation can be matched with the commodity name in the order of the user or the commodity name in the user service list. It is possible to identify whether the user wants to consult the goods that have been purchased or have been consulted or paid attention to, thereby matching the information of the goods, the related orders and the service order to more accurately identify the user's intention.
In some embodiments, the object database comprises a goods number, a service order number, or an order number in the e-commerce platform. And under the condition that the key word is a number, matching the key word with a commodity number, a service order number or an order number in the object database according to a numbering rule, and determining at least one of a commodity, an order and a service order corresponding to the key word.
Different numbering types correspond to different numbering rules, different regular expressions can be set, the numbers in the key participles are matched with the regular expressions, the type to which the numbers belong is determined, and then the numbers are matched with the numbers of the corresponding types in the object database. So as to determine the goods, orders or service orders corresponding to the key participles through the numbers. When the commodity is matched through the serial number, whether the commodity is in the order or the service order of the user can be matched, and the order or the service order corresponding to the key participle is further determined. When the serial number cannot be accurately matched with the corresponding object, fuzzy matching can be performed on the serial number and the serial number in the object database according to a preset rule, and the matched objects with the preset number are used as the objects corresponding to the serial number.
In some embodiments, the object database includes titles of items in the e-commerce platform. And matching the key word with the title of the commodity in the object database, and identifying at least one item of the commodity, the order and the service list corresponding to the key word.
The title of the commodity can be subjected to word segmentation, the key word segmentation is compared with the title word segmentation, and an object corresponding to the key word segmentation is determined. Under the condition that the commodity is matched through the key word segmentation, whether the commodity is in the order or the service order of the user can be matched, and then the order or the service order corresponding to the key word segmentation is determined. When the serial number cannot be accurately matched with the corresponding object, fuzzy matching can be performed on the serial number and the serial number in the object database according to a preset rule, and the matched objects with the preset number are used as the objects corresponding to the serial number.
In some embodiments, the object database includes specific words, such as words of goods, orders, services, and the like, that may represent object category information. And under the condition that the specific object cannot be matched and the key word is a character, matching the key word with a specific word which represents object category information in the object database to determine the category of the object. In the case where the object type is a commodity, a commodity which is interested or purchased by the user or a commodity sold hot can be used as an object corresponding to the key participle. The commodities which the user is interested in include commodities which the user focuses on, commodities which the browsing times of the user exceed a threshold value, commodities which the user browses at the last time, and the like. The hot sold commodities comprise commodities of which the sales volume exceeds the corresponding threshold value within the preset time. And when the object type is an order, taking the order of the user as an object corresponding to the key word segmentation. And when the object type is the service list, taking the service list of the user as the object corresponding to the key word segmentation.
In some embodiments, multiple key segments are matched to the same object, for example, the same item is matched by the item name and the item number, and the results can be merged to remove repeated objects, so that the object to be consulted by the user can be more accurately determined.
In step S106, the description information of the object is provided to the user so that the user confirms that the object to be consulted is the object.
The description information of the object includes at least one item of information of the name, the picture, the corresponding commodity number, the corresponding order number, the corresponding service order number and the state of the object. The state of the object includes, for example: during the distribution, the state is completed. The description information of the object can be provided for the user in the form of a BOOL card type pop-up window, and options are provided for the user to confirm whether the object to be consulted is the object.
In some embodiments, the object may be matched to objects in the user consultation context; providing the description information of the object to the user in the case of matching to the same object; and under the condition that the same object is not matched, inquiring the user whether the object to be consulted is the object or not, and under the condition that the user confirms that the object to be consulted is the object, providing the description information of the object to the user. If the user consultation record is already available, the probability that the user wants to consult the object is higher, and the description information of the object can be directly provided for the user. The object in the user consultation context can be recorded and updated in real time, and can be an object obtained by identifying pictures, a character input by the user or information sent by the user in other forms (such as links).
In some embodiments, in the case where there are a plurality of objects, the respective objects are provided to the user in the form of a list; and acquiring the object to be consulted selected by the user from the list, and providing the description information of the object to be consulted to the user. The object list can be displayed to the user through the BOOL card, and the user can select one or more objects as the objects to be consulted by clicking.
Under the condition that characters are matched with commodity names, multiple commodities can be matched, or under the condition that codes are subjected to fuzzy matching, multiple commodities can be matched, or under the condition that specific commodities cannot be matched accurately and multiple related commodities can not be matched. The goods of interest to the user or purchased goods or hot-sold goods can be selected and provided to the user as objects for the user to select one or more of the goods for consultation. The commodities which the user is interested in include commodities which the user focuses on, commodities which the browsing times of the user exceed a threshold value, commodities which the user browses at the last time, and the like. The hot sold commodities comprise commodities of which the sales volume exceeds the corresponding threshold value within the preset time. The user may be provided with a list of items including a preset number of items. The commodity list can be constructed by selecting a preset number of commodities according to the sequence of the interest degrees of the users from large to small or the sequence of the hot sales degrees of the commodities from large to small.
The method of the embodiment identifies the characters in the picture after receiving the picture sent by the user in the customer service process, identifies the object corresponding to the characters through matching of the characters and the object database, and further can provide the description information of the object for the user so that the user can confirm that the object to be consulted is the object. The method of the embodiment can intelligently identify the user intention, determine the object to be consulted by the user and make a corresponding response under the condition that the user sends the picture. The method of the embodiment enables the user to send the picture instead of manual input, facilitates consultation of the user, improves service quality of the user, and improves user experience.
By matching the characters with the object database, specific objects may not be matched. The object may be determined using page layout information of a picture transmitted by a user. The object database includes page layout information for representing the layout of pages corresponding to different types of objects. For example, the page layouts of the item detail page, the order page, and the service sheet are all different. The screenshots sent by the user correspond to different page layout information, and the OCR technology can restore layout information, namely page layout information, so that the object type can be determined by matching the page layout information. When the object type is a commodity, the commodity which is interested or purchased by the user can be used as the object corresponding to the key participle. The commodities which the user is interested in include commodities which the user focuses on, commodities which the browsing times of the user exceed a threshold value, commodities which the user browses at the last time, and the like. And when the object type is an order, taking the order of the user as an object corresponding to the key word segmentation. And when the object type is the service list, taking the service list of the user as the object corresponding to the key word segmentation.
Further embodiments of the customer service method of the present invention are described below in conjunction with fig. 2.
FIG. 2 is a flow chart of further embodiments of a customer service method of the present invention. As shown in fig. 2, the method of this embodiment includes: steps S202 to S218.
In step S202, a picture sent by the user during the customer service is received.
In step S204, characters in the picture are recognized using an OCR algorithm.
In step S206, the characters are segmented, stop words are filtered, and wrong characters are corrected to obtain key segments.
In step S208, it is identified whether the key segmented word includes a sensitive word, if so, step S209 is executed, otherwise, step S210 is executed.
In step S209, a sensitive word response is made to the user.
For example, prompting the user to switch to a manual customer service and providing a manual customer service link. "C (B)
In step S210, the character is matched with the object database, and an object corresponding to the character is identified.
In step S212, the number of recognized objects is determined. If no object is recognized, step S213 is executed, if a plurality of objects are recognized, step S214 is executed, and if one object is recognized, step S216 is executed.
In step S213, the user is replied to in accordance with the preset reply mode.
For example, the user is prompted that the picture cannot be identified and the text is input to describe the question to be consulted.
In step S214, the respective objects are provided to the user in the form of a list.
The user may be prompted to select one or more of the objects for consultation.
In step S215, information of an object to be consulted selected from the list by the user is received. Step S218 is performed.
In step S216, the object is matched with the object in the user consultation context, and it is determined whether the object exists in the context, if so, step S218 is performed, otherwise, step S217 is performed.
In step S217, the user is asked whether the object to be consulted is the object, and if so, step S218 is performed, otherwise, step S213 is performed.
In step S218, the description information of the object is provided to the user.
While the user may be prompted to ask any questions about the object.
The present invention also provides a customer service device, described below in conjunction with fig. 3.
FIG. 3 is a block diagram of some embodiments of a customer service device of the present invention. As shown in fig. 3, the apparatus 30 of this embodiment includes: a character recognition module 302, an object recognition module 304, and an information transmission module 306.
A character recognition module 302 for recognizing characters in pictures sent by a user during a customer service.
In some embodiments, the text in the picture is recognized according to an optical character recognition algorithm.
And the object identification module 304 is used for matching the characters with an object database and identifying the objects corresponding to the characters, wherein the object database comprises object information in the e-commerce platform.
In some embodiments, the object recognition module 304 is configured to perform word segmentation on the characters, filter stop words, and correct wrong characters to obtain key word segments; and matching the key word segmentation with the object in the object database, and identifying the object corresponding to the key word segmentation.
Further, in some embodiments, the object recognition module 304 is configured to input the segmented words after the filtering of the stop words into a pre-trained error correction model to obtain a correct probability of each segmented word, use the segmented words with the correct probability lower than a threshold as error segmented words, replace the error segmented words with similar shapes, re-input the error correction model and determine the correct probability of each segmented word until the segmented word with the highest correct probability is selected as the key segmented word. Or the object recognition module 304 is configured to match the segmented words after the stop words are filtered with the wrong word bank, determine the wrong segmented words, and correct the wrong segmented words.
Further, in some embodiments, the object database includes at least one of a name of the good in the e-commerce platform, a name of the good in the user order, and a name of the good in the user service order. The object identification module 304 is configured to, when the key word is a character, match the key word with a name of a commodity in the object database, and identify at least one of the commodity, the order, and the service order corresponding to the key word.
In some embodiments, the object database comprises a goods number, a service order number, or an order number in the e-commerce platform; the object identification module 304 is configured to, when the key word is a number, match the key word with a product number, a service order number, or an order number in the object database according to a numbering rule, and determine at least one of a product, an order, and a service order corresponding to the key word.
The information sending module 306 is configured to provide the description information of the object to the user, so that the user can confirm that the object to be consulted is the object.
In some embodiments, the information sending module 306 is used to match the object with an object in the user consultation context; providing the description information of the object to the user in the case of matching to the same object; and under the condition that the same object is not matched, inquiring the user whether the object to be consulted is the object or not, and under the condition that the user confirms that the object to be consulted is the object, providing the description information of the object to the user.
In some embodiments, the information sending module 306 is configured to provide each object to the user in a list if the object is multiple; and acquiring the object to be consulted selected by the user from the list, and providing the description information of the object to be consulted to the user.
In some embodiments, the description information of the object includes at least one item of information of a name, a picture, a corresponding commodity number, a corresponding order number, and a corresponding service order number of the object.
The customer service devices in embodiments of the invention may each be implemented by a variety of computing devices or computer systems, as described below in conjunction with fig. 4 and 5.
FIG. 4 is a block diagram of some embodiments of a customer service device of the present invention. As shown in fig. 4, the apparatus 40 of this embodiment includes: a memory 410 and a processor 420 coupled to the memory 410, the processor 420 configured to perform a customer care method in any of the embodiments of the invention based on instructions stored in the memory 410.
FIG. 5 is a block diagram of another embodiment of a customer service device of the present invention. As shown in fig. 2, the apparatus 50 of this embodiment includes: memory 510 and processor 520 are similar to memory 410 and processor 420, respectively. An input output interface 530, a network interface 540, a storage interface 550, and the like may also be included. These interfaces 530, 540, 550 and the connections between the memory 510 and the processor 520 may be, for example, via a bus 560. The input/output interface 530 provides a connection interface for input/output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 540 provides a connection interface for various networking devices, such as a database server or a cloud storage server. The storage interface 550 provides a connection interface for external storage devices such as an SD card and a usb disk.
The present invention also provides a computer-readable storage medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the customer service method of any of the preceding embodiments.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A customer service method, comprising:
identifying characters in a picture sent by a user in a customer service process;
matching the characters with an object database, and identifying objects corresponding to the characters, wherein the object database comprises object information in an e-commerce platform;
providing the description information of the object to the user so that the user can confirm that the object to be consulted is the object;
wherein the matching the characters with an object database, and the identifying the objects corresponding to the characters comprises:
performing word segmentation on the characters, filtering stop words and correcting wrong characters to obtain key word segmentation;
matching the key word with the commodity name in the object database under the condition that the key word is a character, and identifying at least one item of the commodity, the order and the service list corresponding to the key word;
under the condition that the key participle is a serial number, the commodity serial number, the service order serial number or the order serial number in the object database comprising the electronic commerce platform, matching the key word with the commodity number, the service order number or the order number in the object database according to the numbering rule, determining at least one item of the commodity, the order and the service order corresponding to the key word, the commodity number, the service order number and the order number belong to different number types, the different number types correspond to different number rules, different regular expressions are set, the number in the key participle is matched with the regular expressions to determine the number type to which the number belongs, and the number is matched with the number of the corresponding number type in the object database so as to determine the commodity, the order or the service order corresponding to the key participle through the number;
wherein correcting the erroneous character comprises:
inputting the participles after the stop words are filtered into a pre-trained error correction model to obtain the correct probability of each participle, taking the participles with the correct probability lower than a threshold value as error participles, replacing the error participles with words with similar shapes, re-inputting the error correction model and determining the correct probability of each participle until the participles with the highest correct probability are selected as key participles;
or matching the participles after the stop words are filtered with a wrong word bank, determining wrong participles and correcting the wrong participles.
2. The customer service method of claim 1,
the providing the description information of the object to the user comprises:
matching the object with an object in the user consultation context;
providing the description information of the object to a user in the case of matching to the same object;
and inquiring the user whether the object to be consulted is the object or not under the condition that the same object is not matched, and providing the description information of the object to the user under the condition that the user confirms that the object to be consulted is the object.
3. The customer service method of claim 1,
the providing the description information of the object to the user comprises:
providing each object to the user in a list form if the object is multiple;
and acquiring an object to be consulted selected from the list by the user, and providing the description information of the object to be consulted to the user.
4. The customer service method according to any one of claims 1 to 3,
the characters in the picture are identified according to an optical character identification algorithm;
or the description information of the object comprises at least one item of information of the name, the picture, the corresponding commodity number, the corresponding order number and the corresponding service order number of the object.
5. A customer service device comprising:
the character recognition module is used for recognizing characters in pictures sent by a user in the customer service process;
the object identification module is used for matching the characters with an object database and identifying the objects corresponding to the characters, wherein the object database comprises object information in an electronic commerce platform;
the information sending module is used for providing the description information of the object to the user so that the user can confirm that the object to be consulted is the object;
the object recognition module is used for segmenting the characters, filtering stop words and correcting wrong characters to obtain key segmentation words; matching the key participle with the commodity name in the object database under the condition that the key participle is a character, identifying at least one of a commodity, an order and a service order corresponding to the key participle, matching the key participle with the commodity name in the object database under the condition that the key participle is a character, determining at least one of a commodity, an order and a service order corresponding to the key participle according to a numbering rule and the commodity number, the service order or the order number in the object database under the condition that the key participle is a number, wherein the commodity number, the service order number and the order number belong to different numbering types, different numbering types correspond to different numbering rules, different regular expressions are set, the numbers in the key participles are matched with the regular expressions, the numbering types to which the numbers belong are determined, the numbers are matched with the numbers of the corresponding numbering types in the object database, and therefore commodities, orders or service lists corresponding to the key participles are determined through the numbers;
the object recognition module is used for inputting the participles after the stop words are filtered into a pre-trained error correction model to obtain the correct probability of each participle, taking the participles with the correct probability lower than a threshold value as error participles, replacing the error participles with words with similar shapes, re-inputting the error correction model and determining the correct probability of each participle until the participles with the highest correct probability are selected as key participles;
or the object identification module is used for matching the segmented words after the stop words are filtered with the wrong word bank, determining the wrong segmented words and correcting the wrong segmented words.
6. The customer service device of claim 5,
the information sending module is used for matching the object with an object in the user consultation context; providing the description information of the object to a user in the case of matching to the same object; and inquiring the user whether the object to be consulted is the object or not under the condition that the same object is not matched, and providing the description information of the object to the user under the condition that the user confirms that the object to be consulted is the object.
7. The customer service device of claim 5,
the information sending module is used for providing each object to the user in a list form under the condition that the number of the objects is multiple; and acquiring an object to be consulted selected from the list by the user, and providing the description information of the object to be consulted to the user.
8. The customer service device of any of claims 5-7,
the characters in the picture are identified according to an optical character identification algorithm;
or the description information of the object comprises at least one item of information of the name, the picture, the corresponding commodity number, the corresponding order number and the corresponding service order number of the object.
9. A customer service device comprising:
a memory; and
a processor coupled to the memory, the processor configured to execute the customer service method of any of claims 1-4 based on instructions stored in the memory device.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810824060.3A CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810824060.3A CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108984777A CN108984777A (en) | 2018-12-11 |
| CN108984777B true CN108984777B (en) | 2021-04-30 |
Family
ID=64550992
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810824060.3A Active CN108984777B (en) | 2018-07-25 | 2018-07-25 | Customer service method, apparatus and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108984777B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111325652A (en) * | 2018-12-17 | 2020-06-23 | 北京京东尚科信息技术有限公司 | Information embedding method, information extracting method, information embedding device, information extracting device, information embedding equipment and computer readable storage medium |
| CN111985945A (en) * | 2019-05-21 | 2020-11-24 | 阿里巴巴集团控股有限公司 | Order determining method and device |
| CN110543812A (en) * | 2019-07-19 | 2019-12-06 | 拉扎斯网络科技（上海）有限公司 | information extraction method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514547A (en) * | 2013-03-29 | 2014-01-15 | 洛阳中创网络技术服务有限公司 | Online customer service communication method and system for enterprise website |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107657230A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank self-help robot character recognition device |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
| US10275522B1 (en) * | 2015-06-11 | 2019-04-30 | State Farm Mutual Automobile Insurance Company | Speech recognition for providing assistance during customer interaction |
-
2018
- 2018-07-25 CN CN201810824060.3A patent/CN108984777B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103514547A (en) * | 2013-03-29 | 2014-01-15 | 洛阳中创网络技术服务有限公司 | Online customer service communication method and system for enterprise website |
| US10275522B1 (en) * | 2015-06-11 | 2019-04-30 | State Farm Mutual Automobile Insurance Company | Speech recognition for providing assistance during customer interaction |
| CN107609056A (en) * | 2017-08-25 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | A kind of question and answer processing method and equipment based on picture recognition |
| CN107657230A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank self-help robot character recognition device |
| CN107807960A (en) * | 2017-09-30 | 2018-03-16 | 平安科技（深圳）有限公司 | Intelligent customer service method, electronic installation and computer-readable recording medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108984777A (en) | 2018-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2019261735B2 (en) | System and method for recommending automation solutions for technology infrastructure issues | |
| CN105426356B (en) | A kind of target information recognition methods and device | |
| CN112070138B (en) | Construction method of multi-label mixed classification model, news classification method and system | |
| CN111639516B (en) | Analysis platform based on machine learning | |
| CN108984777B (en) | Customer service method, apparatus and computer-readable storage medium | |
| CN111858843B (en) | Text classification method and device | |
| CN103678460B (en) | For identifying the method and system for the non-text elements for being suitable to be communicated in multi-language environment | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| US11023720B1 (en) | Document parsing using multistage machine learning | |
| CN111680165A (en) | Information matching method and device, readable storage medium and electronic equipment | |
| EP3819846A1 (en) | Content alignment | |
| CN115293332A (en) | Method, device and equipment for training graph neural network and storage medium | |
| CN113590678A (en) | Portrait analysis method based on internet finance and big data analysis server | |
| EP4283496A1 (en) | Techniques for automatic filling of an input form to generate a listing | |
| CN111858942A (en) | Text extraction method and device, storage medium and electronic equipment | |
| CN113065329A (en) | Data processing method and device | |
| CN111143533A (en) | Customer service method and system based on user behavior data | |
| CN115718889A (en) | Industry classification method and device for company profile | |
| CN115659961A (en) | Method, apparatus and computer storage medium for extracting text viewpoints | |
| CN116127188A (en) | Target feedback value determining method and device, electronic equipment and storage medium | |
| CN115080864A (en) | Artificial intelligence based product recommendation method and device, computer equipment and medium | |
| Hoiriyah et al. | Lexicon-Based and Naive Bayes Sentiment Analysis for Recommending the Best Marketplace Selection as a Marketing Strategy for MSMEs | |
| CN111782601A (en) | Electronic file processing method and device, electronic equipment and machine readable medium | |
| CN115860007B (en) | Method and device for calculating index influence degree, storage medium and electronic equipment | |
| CN112445910B (en) | Information classification method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210527 Address after: 100176 room 1004, 10th floor, building 1, 18 Kechuang 11th Street, Beijing Economic and Technological Development Zone, Daxing District, Beijing Patentee after: Beijing Huijun Technology Co.,Ltd. Address before: 100195 Beijing Haidian Xingshikou Road 65 West Cedar Creative Garden 4 District 11 Building East 1-4 Floor West 1-4 Floor Patentee before: BEIJING JINGDONG SHANGKE INFORMATION TECHNOLOGY Co.,Ltd. Patentee before: BEIJING JINGDONG CENTURY TRADING Co.,Ltd. |
|
| TR01 | Transfer of patent right |