Packaging material consultation service system

Abstract

The invention relates to the technical field of online information consultation, in particular to a packaging material consultation service system which comprises a user side, a server side and a storage side, historical consultation information is analyzed, vocabulary elements in the consultation information are divided into multiple information dimensions, then the consultation information of a user is decomposed into first-level information and second-level information, and the first-level information and the second-level information are stored in the server side. The first-level information and the second-level information are respectively retrieved and calculated in the storage end, the consultation intention value of the user is obtained, the grade of the user is marked according to the consultation intention value, and then the service matching module distributes corresponding exclusive customer service to the user according to the grade of the user. On one hand, the utilization rate of customer service resources is optimized, and on the other hand, the exclusive customer service is allocated to the user according to the user level, so that the transaction volume of user consultation is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of online information consulting, and particularly relates to a packaging material consulting service system. BACKGROUND

[0002] With the increasing competition of market environment, the demand of enterprises for packaging is no longer limited to the traditional protection and containing functions, but develops towards the direction of more individualization, environmental protection and high efficiency.

[0003] The prior art CN116739294A discloses an enterprise management consulting service system based on big data, comprising a data acquisition unit, a cloud database, a user consulting state positioning unit, a service providing state positioning unit, a consulting service bidirectional matching unit, a data summary remodeling unit and a display terminal. The present application provides the best service personnel for the consultation of each enterprise customer in the enterprise management consulting service system through the collection, integration and analysis of data;

[0004] However, in the consulting service system, the number of consulting customers is large, and the types of customers are various, and the intention degree of customers in the process of consulting packaging materials is also different, so the service process of the existing consulting service system in the process of serving the consulting customers only serves the consulting customers according to the set program, and the service process of the intention customers in the consulting customers is not targeted, which easily leads to the loss of intention customers, and further reduces the customer transaction volume in the consulting service system. SUMMARY

[0005] The present application relates to the technical field of online information consulting, and particularly relates to a packaging material consulting service system.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme:

[0007] A packaging material consulting service system, comprising a user end, a service end, a storage end;

[0008] The user end is used for the user to fill in the consulting information and collect the basic information of the user in the consulting service system, and transmits the consulting information and the basic information to the service end, and the basic information includes the user ID and the user identity;

[0009] The storage end is used for storing the consulting related information, wherein the consulting related information includes the packaging material information, the historical consulting user and the basic information thereof, and the historical consulting information, and the storage end is bidirectionally connected with the service end;

[0010] The service end is used for processing the consulting information, wherein the service end comprises a semantic processing module, a historical information grading module, an intention analysis module and a service matching module;

[0011] The semantic processing module is configured to perform word sense resolution on the consultation information to obtain a plurality of lexical elements, and then divide the lexical elements into primary information and secondary information according to a semantic position correspondence table, wherein the semantic position correspondence table refers to a component correspondence position set according to a sentence grammatical structure;

[0012] The historical information grading module is configured to perform deduplication on the primary information in the historical consultation information, and obtain the secondary information in the same primary information, and then perform merging on the secondary information to obtain supplementary information, and then calculate the lexical class values of the lexical elements in the supplementary information, compare the lexical class values with a similar threshold interval, and divide the lexical elements into a plurality of information dimensions according to the comparison result.

[0013] The intention analysis module is configured to obtain the primary information and the secondary information of the user, and then perform retrieval on the primary information and the secondary information respectively to obtain information repetition values and information dimensions of the lexical elements, and then calculate the consultation intention values by using the information repetition values and the information dimensions, and then divide the user into different grades according to the consultation intention values.

[0014] The service matching module is configured to match the user with a dedicated customer service of the same grade according to the grade of the user.

[0015] As a further scheme of the present application, the storage end is provided with a material information storage module, a user information storage module and a consultation information storage module.

[0016] The material information storage module is configured to store the packaging material information.

[0017] The user information storage module is configured to store the basic information of the historical consultation user.

[0018] The consultation information storage module is configured to store the consultation information of the user.

[0019] As a further scheme of the present application, the specific processing method of the semantic processing module includes:

[0020] The semantic position correspondence table is set in the semantic processing module, wherein the semantic position correspondence table refers to a component correspondence position set according to a sentence grammatical structure, and includes a primary structure position and a secondary structure position, the primary structure position refers to the basic structure of the sentence, and the secondary structure position refers to the words for supplementing the primary structure position.

[0021] The consultation information is subjected to word sense resolution, that is, the consultation information is split according to conventional word groups or single characters, and the split conventional word groups or single characters are marked as lexical elements.

[0022] By means of a natural language processing tool, the lexical elements are subjected to component recognition, and the lexical elements are repositioned according to the component reference position, and then the consultation information is decomposed into primary information and secondary information.

[0023] As a further scheme of the present application, the primary information refers to a shorthand sentence composed of primary structure information components, and the secondary information refers to limited information composed of lexical elements corresponding to secondary structure.

[0024] As a further scheme of the present application, the processing method in the historical information grading module comprises:

[0025] The primary information in the historical consultation information is extracted and subjected to deduplication processing, and the primary information obtained after the deduplication processing is marked as simplified information, and the deduplication processing refers to deleting the repeated primary information.

[0026] The simplified information is taken as target information, all the secondary information in the target information is obtained, and is marked as supplementary information.

[0027] The lexical elements in the supplementary information are converted into lexical vectors Bi by using an AI algorithm, each lexical element corresponds to a lexical vector, i represents different lexical elements, and i is a positive integer.

[0028] The lexical vectors Bi are subjected to hierarchical processing to obtain a lexical hierarchical value Bg. The lexical hierarchical value Bg is obtained, wherein a and b belong to i, and a≠b.

[0029] When Bg∈c1, it indicates that the lexical element a and the lexical element b are in a parallel relationship, and when , it indicates that the lexical element a and the lexical element b are in a subordinate relationship, and c1 is a similar threshold interval.

[0030] When the lexical element a and the lexical element b are in a subordinate relationship, |Ba| and |Bb| are extracted, respectively, if |Ba|>|Bb|, the lexical element b belongs to the lexical element a, otherwise, if |Ba|<|Bb|, the lexical element a belongs to the lexical element b, that is, the lexical element a is the concrete lexical element of the lexical element b, if |Ba|=|Bb|, an abnormal signal is generated and transmitted to an administrator.

[0031] At this time, the lexical elements in the supplementary information are divided into several information dimensions by the lexical hierarchical value Bg.

[0032] As a further scheme of the present application, the supplementary information is the sum of the secondary information corresponding to the same primary information, and when the secondary information is merged, the repeated lexical elements need to be deleted, and when the primary information is subjected to deduplication processing, the repeated values of the primary information need to be counted to obtain information repeated values.

[0033] As a further scheme of the present application, the processing method of the intention analysis module comprises:

[0034] The first-level information is searched in the information base to obtain an information repetition value CF, meanwhile, the lexical elements in the second-level information are searched in the information dimension, and the information dimension of the lexical elements in the second-level information at this time is marked as WEd, and d represents different lexical elements in the second-level information;

[0035] Adopting An inquiry intention value Yz of the user is obtained, wherein WEm is the maximum value in the information dimension WEd, tx represents the continuous online duration of the user, Ms represents the number of information sent by the user within the continuous online duration tx, H represents a user repurchase coefficient, n represents the repurchase times of the user, K1 and K2 are both proportional factors, and K3 is an identity coefficient determined by the identity of the user;

[0036] According to the inquiry intention value Yz, when YzX1, the user is marked as the first grade, if X1YzX2, the user grade is marked as the second grade, and if YzX2, the user grade is marked as the third grade, X1 and X2 are both threshold values, and X1X2.

[0037] As a further scheme of the present application, the processing method of the service matching module comprises:

[0038] According to the working experience and the order transaction rate of the service staff, the service staff is divided into multiple service grades, and the service grades include a first service grade, a second service grade and a third service grade, wherein the first service grade

[0039] Then, according to the user grade, the corresponding service staff in the corresponding service grade is selected as the exclusive customer service of the corresponding user.

[0040] Compared with the prior art, the present application has the following advantages:

[0041] The application obtains first-level information and second-level information by preprocessing the consultation information of the user, arranges the historical consultation information, deletes repeated first-level information, merges the second-level information corresponding to the repeated first-level information to obtain supplementary information, analyzes each lexical element in the supplementary information to obtain lexical hierarchy values between any two lexical elements, divides the lexical elements in the supplementary information into multiple information dimensions according to the lexical hierarchy values, respectively searches the first-level information and the second-level information according to the consultation information of the user at this time, calculates the search results, obtains the consultation intention value of the user according to the calculation results, marks the user according to the consultation intention value, and then the service matching module allocates the corresponding exclusive customer service to the user according to the user grade, which optimizes the utilization rate of customer service resources on the one hand, and effectively improves the transaction volume of user consultation on the other hand by allocating exclusive customer service to the user according to the user grade. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is a system structure schematic diagram of the application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments.

[0044] Referring to Figure 1 A packaging material consultation service system, comprising a user end, a service end, and a storage end;

[0045] The user end is used for filling in consultation information by the user and transmitting the consultation information to the service end, and is also used for collecting the basic information of the user when the user fills in the consultation information and transmitting the basic information to the service end, wherein the basic information comprises a user ID and a user identity, and the user identity comprises a personal identity and an enterprise identity, the personal identity refers to purchasing packaging materials in a personal identity, and the enterprise identity refers to purchasing packaging materials in the name of an enterprise;

[0046] The storage end is used for storing consultation related information, in an embodiment of the application, according to the function of the storage end, a material information storage module, a user information storage module, and a consultation information storage module are arranged in the storage end;

[0047] The material information storage module is used for storing packaging material information;

[0048] The user information storage module is used for storing the basic information of historical consultation users;

[0049] The consultation information storage module is configured to store the consultation information of the user, and it is to be noted that the consultation information stored herein is the first-level information and the second-level information processed by the semantic processing module, and the consultation information storage module is electrically connected to the historical information grading module in a bidirectional manner;

[0050] The storage end is electrically connected to the service end in a bidirectional manner;

[0051] The service end is configured to receive the basic information and the consultation information of the user, and based on the processing function of the service end, the service end is divided into a semantic processing module, a historical information grading module, an intention analysis module and a service matching module;

[0052] The semantic processing module is configured to pre-process the consultation information based on the received consultation information, and obtain the first-level information and the second-level information according to the pre-processing result, and the method for pre-processing the consultation information includes:

[0053] A semantic position correspondence table is arranged in the semantic processing module, and the semantic position correspondence table is configured to correspond to the positions of components arranged according to the grammatical structure of a sentence, and the positions of components include first-level structure positions and second-level structure positions, the first-level structure positions refer to the basic structure of a sentence, and the second-level structure positions refer to the words used to supplement the first-level structure positions;

[0054] For example, according to the grammatical structure of a sentence, the subject, the predicate and the object will appear in a complete sentence, at this time, the subject, the predicate and the object are set as the first-level structure positions, in addition to the subject-predicate-object, there are also adverbial, complement and attributive grammars to modify the subject-predicate-object, so as to make the description of the sentence more accurate, at this time, the adverbial, the complement and the attributive are set as the second-level structure positions;

[0055] The consultation information is extracted, and the consultation information is subjected to word meaning decomposition, that is, the consultation information is split according to conventional word groups or single characters, and the split conventional word groups or single characters are marked as lexical elements;

[0056] With the aid of a natural language processing tool, the lexical elements are subjected to component recognition, and the lexical elements are repositioned according to the component correspondence positions, so as to decompose the consultation information into the first-level information and the second-level information;

[0057] It is to be noted that the first-level information refers to a shortened sentence composed of only the subject, the predicate and the object, the shortened sentence expresses the demand of the user concisely and accurately, and simplifies the processing steps in the subsequent processing of the consultation information, thereby saving the computing resources, and the second-level information refers to the limited information composed of the adverbial, the complement and the attributive, the limited information is used to further limit and supplement the shortened information, so as to make the information expressed by the shortened sentence more accurate;

[0058] The semantic processing module then transmits both primary and secondary information to the intent analysis module;

[0059] The historical information categorization module divides the consultation information based on historical user inquiries into multiple information dimensions. Specific methods for categorizing consultation information include:

[0060] Extract primary information from historical consultation information, and perform deduplication on the primary information. Mark the primary information obtained after deduplication as simplified information. Deduplication means deleting duplicate primary information.

[0061] It should be further explained that after the consultation information is abbreviated into first-level information, there will be a high degree of repetition in the first-level information. Therefore, it is necessary to deduplicatize the first-level information to avoid repeated data analysis in the future.

[0062] It should be further explained that when deduplicating the primary information, it is necessary to count the duplicate values ​​of the primary information to obtain the duplicate information values.

[0063] Then, the simplified information is used as target information, all secondary information in the target information is obtained, and marked as supplementary information;

[0064] It should be further explained that the supplementary information is the sum of the secondary information corresponding to the same primary information. When merging the secondary information, duplicate word elements need to be deleted. For example, there are 3 identical primary information A, and their corresponding secondary information are (a11, a12, a13), (a21, a22, a23), and (a31, a32, a33). If the primary information A is taken as the target information, the supplementary information of the target information is (a11, a12, a13, a21, a22, a23, a31, a32, a33).

[0065] The vocabulary elements in the supplementary information are converted into vocabulary vectors Bi using an AI algorithm. Each vocabulary element corresponds to a vocabulary vector, where i represents a different vocabulary element and i is a positive integer. In this embodiment, the AI ​​algorithm used is one-hot encoding technology.

[0066] use Obtain the lexical class value Bg, where both a and b belong to i, and a ≠ b;

[0067] When Bg∈c1, it indicates that lexical element a and lexical element b are in a parallel relationship. When, it indicates that word element a and word element b are subordinate to each other, and c1 is the similarity threshold interval. In this embodiment, the similarity threshold interval is set to [0.9, 1].

[0068] When the lexical element a and the lexical element b are in a subordinate relationship, |Ba| and |Bb| are extracted respectively, if |Ba| > |Bb|, the lexical element b belongs to the lexical element a, otherwise, if |Ba| < |Bb|, the lexical element a belongs to the lexical element b, that is, the lexical element a is the concrete lexical of the lexical element b, if |Ba| = |Bb|, an abnormal signal is generated and transmitted to the administrator, and the administrator confirms the abnormal signal;

[0069] The lexical elements in the supplementary information are then divided into a plurality of information dimensions by the lexical hierarchy value Bg, and the information dimensions are arranged in descending order, and the more concrete the lexical elements in the information dimension, the greater the information dimension value;

[0070] The information dimensions are then transmitted to the depth comparison module by the historical information division module;

[0071] The intention analysis module analyzes the consultation intention value of the user according to the received primary information and secondary information, and the specific analysis method includes:

[0072] The primary information is extracted and searched, and the information repetition value CF is obtained, then the lexical elements in the secondary information are searched in the information dimension, and the information dimension of the lexical elements in the secondary information at this time is marked as WEd, d represents different lexical elements in the secondary information;

[0073] The consultation intention value Yz of the user is obtained, where WEm is the maximum value in the information dimension WEd, tx represents the continuous online duration of the user, Ms represents the number of information sent by the user within the continuous online duration tx, H represents the user repurchase coefficient, which is valued by a person skilled in the art, n represents the user repurchase times, if the user is the first time to purchase, n is 0, K1 and K2 are proportional factors, and their specific values are set by a person skilled in the art, and K3 is an identity coefficient, which is determined by the user identity;

[0074] According to the consultation intention value Yz, when Yz < X1, the user is marked as the first level, if X1 ≤ Yz < X2, the user level is marked as the second level, if Yz ≥ X2, the user level is marked as the third level, X1 and X2 are threshold values, and X1 < X2, and the specific values of X1 and X2 are set by a person skilled in the art;

[0075] It should be further explained that in the user level, the first level < the second level < the third level;

[0076] The user level is then transmitted to the service matching module by the intention analysis module;

[0077] The service matching module is used to receive the user level, and assign the corresponding exclusive customer service to the user according to the user level;

[0078] In another embodiment of the present application, the service matching module divides the service staff into multiple service levels according to the work experience of the service staff and the order transaction rate, and the service levels include a first service level, a second service level and a third service level, wherein the first service level < the second service level < the third service level;

[0079] Then, according to the user level, a corresponding service staff is selected as a corresponding user's exclusive customer service in a corresponding service level, and the user is served through the consultation service system.

[0080] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacements or changes within the technical range disclosed by the present application according to the technical solution and the inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. A packaging material consulting service system, characterized in that, Includes the user end, server end, and storage end; The user terminal is used by users to fill in consultation information and collect basic user information in the consultation service system, and transmits the consultation information and basic information to the server. The basic information includes user ID and user identity. The storage end is used to store consultation-related information, including packaging material information, historical consultation users and their basic information, and historical consultation information. The storage end and the server are bidirectionally connected. The server-side component processes consultation information and includes a semantic processing module, a historical information grading module, an intention analysis module, and a service matching module. The semantic processing module is used to decompose the consultation information into multiple lexical elements. Then, according to the semantic position lookup table, the lexical elements are divided into primary information and secondary information. The semantic position lookup table refers to the component reference position set according to the sentence grammatical structure. The historical information grading module is used to first deduplicate the first-level information in the historical consultation information, and at the same time obtain the second-level information in the same first-level information. The second-level information is merged to obtain supplementary information. Then, the word elements in the supplementary information are calculated to obtain the word class value. The word class value is compared with the similarity threshold range, and the word elements are divided into multiple information dimensions according to the comparison results. The intent analysis module is used to obtain the user's primary and secondary information, and to retrieve the primary and secondary information separately to obtain the information repetition value and the information dimension of the vocabulary elements. The consultation intent value is calculated using the information repetition value and information dimension, and then the user is classified into different levels based on the consultation intent value. The service matching module is used to match users with dedicated customer service representatives of the same level based on their user level.

2. The packaging material consulting service system according to claim 1, characterized in that, The storage terminal is equipped with a material information storage module, a user information storage module, and a consultation information storage module; The material information storage module is used to store packaging material information; The user information storage module is used to store basic information of historical consulting users; The consultation information storage module is used to store users' consultation information.

3. The packaging material consulting service system according to claim 1, characterized in that, The specific processing methods of the semantic processing module include: In the semantic processing module, a semantic position lookup table is set up. The semantic position lookup table refers to the component lookup positions set according to the sentence grammatical structure, including primary structure positions and secondary structure positions. Primary structure positions refer to the basic structure of the sentence, and secondary structure positions refer to words that provide supplementary explanations for the primary structure positions. The consultation information is decomposed into semantics, which means breaking down the consultation information into regular phrases or individual characters, and marking the decomposed regular phrases or individual characters as lexical elements. By using natural language processing tools, the components of lexical elements are identified, and the lexical elements are repositioned according to their corresponding positions, thereby decomposing the consultation information into primary and secondary information.

4. The packaging material consulting service system according to claim 3, characterized in that, Primary information refers to abbreviated statements whose information components are composed of primary structures, while secondary information refers to limiting information composed of lexical elements corresponding to secondary structures.

5. The packaging material consulting service system according to claim 1, characterized in that, The processing methods in the historical information tiering module include: Extract primary information from historical consultation information and perform deduplication. Mark the primary information obtained after deduplication as simplified information. Deduplication means deleting duplicate primary information. Simplified information is used as target information, all secondary information in the target information is obtained and marked as supplementary information; The vocabulary elements in the supplementary information are transformed into vocabulary vectors Bi using an AI algorithm. Each vocabulary element corresponds to a vocabulary vector, where i represents a different vocabulary element and i is a positive integer. use Obtain the lexical class value Bg, where both a and b belong to i, and a ≠ b; When Bg∈c1, it indicates that lexical element a and lexical element b are in a parallel relationship. When c1 is a similarity threshold interval, it indicates that word element a and word element b are subordinate to each other. When vocabulary element a and vocabulary element b are subordinate to each other, extract |Ba| and |Bb| respectively. If |Ba| > |Bb|, then vocabulary element b belongs to vocabulary element a. Conversely, if |Ba| < |Bb|, then vocabulary element a belongs to vocabulary element b. That is, vocabulary element a is a concrete vocabulary of vocabulary element b. If |Ba| = |Bb|, an abnormal signal is generated and transmitted to the administrator. At this point, the lexical elements in the supplementary information are divided into several information dimensions by the lexical class value Bg.

6. The packaging material consulting service system according to claim 5, characterized in that, The supplementary information is the sum of the secondary information corresponding to the same primary information. When merging the secondary information, duplicate word elements need to be deleted. When deduplicating the primary information, the duplicate values ​​of the primary information need to be counted to obtain the information duplication value.

7. A packaging material consulting service system according to claim 6, characterized in that, The processing methods of the intention analysis module include: Search the primary information in the information database to obtain the information duplication value CF. At the same time, search the vocabulary elements in the secondary information in the information dimension and mark the information dimension of the vocabulary elements in the secondary information as WEd, where d represents different vocabulary elements in the secondary information. use The user's consultation intention value Yz is obtained, where WEm is the maximum value in the information dimension WEd, tx represents the user's continuous online time, Ms represents the number of messages sent by the user within the continuous online time tx, H represents the user repurchase coefficient, n represents the number of user repurchases, K1 and K2 are both proportional factors, and K3 is the identity coefficient, which is determined by the user's identity. Based on the consultation intention value Yz, when Yz < X1, the user is marked as the first level; if X1 ≤ Yz < X2, the user is marked as the second level; if Yz ≥ X2, the user is marked as the third level. X1 and X2 are both thresholds, and X1 < X2.

8. The packaging material consulting service system according to claim 1, characterized in that, The processing methods of the service matching module include: Based on the work experience and order completion rate of service staff, service staff are divided into multiple service levels, including first service level, second service level and third service level, where first service level < second service level < third service level; Then, based on the user's level, select the corresponding service staff as the user's dedicated customer service representative.

Images