CN110232589B - Intention customer analysis system based on big data - Google Patents
Intention customer analysis system based on big data Download PDFInfo
- Publication number
- CN110232589B CN110232589B CN201910407588.5A CN201910407588A CN110232589B CN 110232589 B CN110232589 B CN 110232589B CN 201910407588 A CN201910407588 A CN 201910407588A CN 110232589 B CN110232589 B CN 110232589B
- Authority
- CN
- China
- Prior art keywords
- commodity
- commodities
- intention
- function
- function set
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The invention discloses an intention customer analysis system based on big data, which comprises a client end in communication connection with the system, a commodity classification module, a customer behavior acquisition module, a commodity intention grading module, a keyword extraction module, a commodity function extraction module and a storage module. The commodity intention grading module divides the intention of the customer into two intention grades, can more accurately determine the purchasing intention of the customer, and can provide data support for later-period sales. The keyword extraction module generates a necessary function keyword, an irrelevant function keyword, and a preferred function keyword to represent the functions that the customer most pays attention to for a product of a category, and the functions of the product that the customer does not pay attention to further improve the evaluation of the product. The method realizes further mining of commodities with different intentions, and can be used for advertisement pushing and consumption prediction in the later period.
Description
Technical Field
The invention relates to the technical field of data mining, in particular to an intention customer analysis system based on big data.
Background
On-line shopping is increasingly becoming the mainstream of shopping, a large amount of data can be generated when a customer browses commodities, and how to mine the data to help sales is a hotspot of the current market, but in the prior art, for example, a file with the application number of CN201811278267.1 discloses a customer intention supervision and prediction method based on customer attributes and marketing data, which comprises knowledge acquisition, knowledge base updating and customer purchase intention prediction. However, this method is only to determine whether a customer has an intention to purchase a commodity, and does not perform deep mining on why a user has selected the commodity, and the depth of data mining is low and the degree of visualization is not high.
Disclosure of Invention
In order to solve the above-mentioned technical problems, an object of the present invention is to provide an intention customer analysis system based on big data, which is capable of mining browsing behaviors of customers, further subdividing the intentions of the customers, further mining according to the subdivided results, acquiring functions necessary for the customers to select and purchase a commodity, functions not paying attention to the commodity, and functions as an addendum item, and providing data support for later sales.
The technical problem to be solved by the invention is as follows:
(1) how to deeply mine the browsing behavior of the client.
The purpose of the invention can be realized by the following technical scheme:
an intention customer analysis system based on big data comprises a client end in communication connection with the system, wherein the client end is used for generating a browsing record containing a commodity name and browsing duration of a commodity page after a customer clicks a commodity link, and the system further comprises a commodity classification module, a customer behavior acquisition module, a commodity intention grading module, a keyword extraction module, a commodity function extraction module and a storage module;
the data stored by the storage module comprises commodity classification data, a commodity function database of each commodity, browsing records of users, commodity lists in shopping carts and finished order information;
the commodity classification module is used for classifying all commodities according to a commodity classification table and dividing the commodities according to the same class;
the client behavior acquisition module is used for acquiring browsing records, commodity lists and order information of a client from a client;
the commodity function extraction module acquires the functions of commodities from the commodity function databases of various categories and generates a function set of the commodities;
after analyzing the browsing records of the client and the shopping cart list, the commodity intention grading module divides the browsed commodities into a first-level intention and a second-level intention;
the commodity intention grading module specifically grades commodities according to the following steps:
s1, the commodity intention grading module acquires the commodity name and the browsing duration of the commodity page in each browsing record of the client, and deletes the browsing record of which the browsing duration of the commodity page is less than 3 seconds; then obtaining the number of browsing records with the same commodity name as the browsing times, finally deleting the browsing records with the browsing times smaller than 2, and obtaining the commodity names in the rest browsing records as the first-level intention commodities of the client;
and S2, the commodity intention grading module calculates the browsing times of the corresponding first-level intention commodity from the rest browsing records as real times, compares the real times with a preset comparison value A, if the real times are larger than the comparison value A, upgrades the first-level intention commodity into a second-level intention commodity, and simultaneously takes the commodity in the commodity list in the shopping cart as the second-level intention commodity.
The keyword extraction module is used for analyzing and comparing the function sets of commodities with different intention grades after the commodity intention grading module grades, and screening necessary function keywords, irrelevant function keywords and preferred function keywords of the customers for the functions of the commodities of the class.
Further, the numerical value of the control value a is obtained by the following method:
screening purchased commodities from the completed order information of the client to be used as reference commodities, screening browsing times corresponding to the reference commodities and the number of the reference commodities from browsing records of the client, calculating average browsing times, and taking the average browsing times as a comparison value A, wherein the calculation formula of the average browsing times is as follows:
T＝(t1+t2+…+tn)/n；
where T is the average number of views, (T1+ T2+ … + tn) is the sum of the number of views for each reference item, n is the number of reference items.
Further, the specific steps of the keyword extraction module acquiring the necessary functional keywords, irrelevant functional keywords and preferred functional keywords of the customer for the goods of one category are as follows:
step one, a keyword extraction module randomly selects a class corresponding to a secondary intention commodity, screens out commodities of the class from all the secondary intention commodities as a same-level comparison commodity, screens out all commodities which are the same as the class of the secondary intention commodity from the primary intention commodity as a low-level comparison commodity, and records the number of the same-level comparison commodity and the number of the low-level comparison commodities;
if the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is greater than 1, executing a step two;
if the number of the same-level comparison commodities and the number of the lower-level comparison commodities are equal to 1, executing a third step;
if the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is 0, executing a fourth step;
if the number of the same-level comparison commodities and the number of the low-level comparison commodities are both larger than 1, executing a fifth step;
if the number of the same-level comparison commodities is more than 1 and the number of the lower-level comparison commodities is equal to 1, executing a step six;
if the number of the same-level comparison commodities is more than 1 and the number of the low-level comparison commodities is 0, executing a seventh step;
step two, the keyword extraction module records the function set of the secondary intention commodity as C2The intersection of the functional sets of all low-level control commodities is denoted as JdThe keyword extraction module is according to the formula Fb ═ C2∩JdCalculating necessary function set Fb, and recording the union of all low-level comparison commodity function sets as B by the keyword extraction moduledThen according to the formula Fw ═ Bd-C2Calculating the irrelevant function set Fw, and finally, according to the formula Fy ═ C2-BdCalculating a preferred function set Fy;
thirdly, the keyword extraction module records the function set of the secondary intention commodity as C2The function set of the first-level intention commodity is marked as C1The keyword extraction module is according to the formula Fb ═ C1∩C2Calculating the necessary function set Fb, and then according to the formula Fw ═ C1-C2Calculating the irrelevant function set Fw, and finally, according to the formula Fy ═ C2-C1Calculating a preferred function set Fy;
step four, the keyword extraction module takes the empty set as a necessary function set Fb, an irrelevant function set Fw and an optimal function set Fy;
step five, the keyword extraction module records the intersection of the function sets of all the low-level comparison commodities as JdThe intersection of the function sets of all the same level comparison commodities is recorded as Jt(ii) a The union of the function sets of all the low-level control commodities is denoted as BdThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module is according to the formula Fb ═ Jt∩JdCalculating a necessary function set Fb; then according to the formula Fw ═ Bd-BtCalculating an irrelevant function set Fw, and finally calculating an optimal function set Fy according to a formula Fy-Bd;
step six, the keyword extraction module records the intersection of the function sets of all the same-level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe function set of the first-level intention commodity is marked as C1The keyword extraction module extracts JtAs the necessary function set Fb, according to the formula Fw ═ C1-BtCalculating the irrelevant function set Fw, and finally, according to the formula Fy being Bt-C1Calculating a preferred function set Fy;
step seven, the keyword extraction module records the intersection of the function sets of all the same level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module extracts JtAs the necessary function set Fb, the empty set is used as the irrelevant function set Fw, and finally, B is obtained according to the formula Fyt-JtCalculating a preferred function set Fy;
and step eight, taking the corresponding functions in the necessary function set Fb, the irrelevant function set Fw and the preferred function set Fy obtained in any one of the steps two to seven as the necessary function key, the irrelevant function key and the preferred function key of the customer for the product.
The invention has the beneficial effects that:
(1) the commodity intention grading module is used for simplifying browsing records according to the time length of single browsing and the browsing times, the commodity which is intentionally purchased by a customer is determined, namely the first-level intention commodity, the commodity which is intentionally purchased by the customer is divided into two intention grades by combining the purchasing habit of the customer through the commodity intention grading module, the higher the grade is, the higher the purchasing probability is, after the condition is met, the commodity is upgraded from the first-level intention commodity to the second-level intention commodity, meanwhile, the commodity in the shopping cart is directly brought into the second-level intention commodity, the purchasing intention of the customer can be more accurately determined, and data support can be provided for later-period sales.
(2) The keyword extraction module combines and analyzes the grading result of the commodity intention grading module and the function of the commodity to generate a necessary function keyword, an irrelevant function keyword and a preferred function keyword, so as to represent the most important function of the customer for the commodity of one category, the unconcerned commodity function and further improve the evaluation function of the commodity. The method realizes further mining of commodities with different intentions, and can be used for advertisement pushing and consumption prediction in the later period.
Drawings
The invention will be further described with reference to the accompanying drawings.
FIG. 1 is a system block diagram 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. 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.
Referring to fig. 1, the embodiment provides an intention client analysis system based on big data, which includes a client in communication connection with the system, wherein the client is used for generating a browsing record including a commodity name and a browsing duration of a commodity page after a client clicks a commodity link, and the system further includes a commodity classification module, a client behavior acquisition module, a commodity intention classification module, a keyword extraction module, a commodity function extraction module, and a storage module;
the data stored by the storage module comprises commodity classification data, a commodity function database of each commodity, browsing records of users, commodity lists in shopping carts and finished order information;
the commodity classification module is used for classifying all commodities according to a commodity classification table and dividing the commodities according to the same class; each product type in the product classification table is independent.
The client behavior acquisition module is used for acquiring browsing records, commodity lists and order information of a client from the client and storing the browsing records, the commodity lists and the order information into the storage module;
the commodity function extraction module acquires the functions of the commodities from the commodity function databases of the various categories and generates a function set of the commodities; the commodity function database of each commodity comprises functions of all commodities, the functions of the commodities are collected as a subset of the commodity function database of each commodity, for example, the commodity function database of the washing machine has functions of all commodities under the commodity, such as automatic drying, timed washing, automatic cleaning and the like, and a certain model of washing machine as a commodity has the function of automatic drying, but does not have the functions of timed washing and automatic cleaning. Another type of washing machine has a function of washing laundry at regular time without automatic drying.
After analyzing the browsing records of the client and the shopping cart list, the commodity intention grading module divides the browsed commodities into a first-level intention and a second-level intention; the number of times and duration of browsing the product by the customer may reflect the interest level of the product, but there may be a case of misoperation or uninteresting after browsing, so that the browsing behavior of the customer needs to be mined to determine whether the customer is uninteresting or casual watching the product, or has an intention to purchase.
The commodity intention grading module specifically grades commodities according to the following steps:
s1, because the collected browsing records may have interference items, the interference items need to be eliminated first, especially in the case that the commodity links are clicked by misoperation or the user is not interested after the commodity links are seen. The commodity intention grading module acquires the commodity name and the browsing duration of the commodity page in each browsing record of the client, and deletes the browsing record of which the browsing duration of the commodity page is less than 3 seconds; the browsing is ended very quickly to indicate that the client is not interested or the client clicks by misoperation, so that the browsing is required to be eliminated, the number of browsing records with the same commodity name is acquired as the browsing times, and finally the browsing records with the browsing times smaller than 2 are deleted, namely the commodity page is browsed only once, so that the data reference is not strong, and the commodity names in the rest browsing records are required to be removed to be acquired as the primary intention commodity of the client; the first-level intention commodity is a commodity which is browsed repeatedly and carefully, and indicates that the customer has an intention to purchase.
And S2, the commodity intention grading module calculates the browsing times of the corresponding primary intention commodities from the rest browsing records in the S1 as real times, compares the real times with a preset comparison value A, upgrades the primary intention commodities into secondary intention commodities if the real times are greater than the comparison value A, and takes the commodities in the commodity list in the shopping cart as the secondary intention commodities, namely, the purchase intention of the customers to the commodities is high. When the browsing times exceed the comparison value A, the client is shown to browse the commodity for many times, the interest degree is high, and therefore the intention level of the commodity is upgraded. The addition of a shopping cart is also an action with higher purchasing intention, so that the commodities in the shopping cart need to be included in secondary intention commodities.
The method is characterized in that a user can finish purchasing commodities after browsing for a plurality of times, and some people need to observe for a plurality of times and then place orders, so that different comparison values A are set by analyzing the shopping habits of the user according to different shopping habits of each person to determine the browsing habits of the user when the user purchases the commodities, and the comparison values A can be empirical values or can be determined by a lower method.
The numerical value of the control value A is obtained by the following method:
screening purchased commodities from the completed order information of the client to be used as reference commodities, screening browsing times corresponding to the reference commodities and the number of the reference commodities from browsing records of the client, calculating average browsing times, and taking the average browsing times as a comparison value A, wherein the calculation formula of the average browsing times is as follows:
t ═ T1+ T2+ … + tn)/n, where T is the average number of viewings, (T1+ T2+ … + tn) is the sum of the number of viewings for each reference item, and n is the number of reference items. If there are 3 items in the order information and the browsing times are 4, 5, and 3, respectively, the comparison value a is 4, which indicates that the probability of purchase is high after the customer browses four times.
The keyword extraction module is used for analyzing and comparing the function sets of commodities with different intention grades after the commodity intention grading module grades, and screening necessary function keywords, irrelevant function keywords and preferred function keywords of the customers for the functions of the commodities of the class. The necessary function keywords can be regarded as the most important functions when the customer selects the commodities, only the commodities with the functions are selected, irrelevant function keywords can be regarded as commodity functions which are not concerned by the customer, the evaluation of the customer on the commodities is not influenced if the functions exist, and the preferred function keywords can be regarded as the functions which can further improve the evaluation of the commodities and do not reduce the evaluation of the commodities on the premise that the necessary functions are met.
The specific steps of the keyword extraction module for acquiring the necessary functional keywords, irrelevant functional keywords and preferred functional keywords of the customers for the commodities of one category are as follows:
step one, a keyword extraction module randomly selects a class corresponding to a secondary intention commodity, screens out commodities of the class from all the secondary intention commodities as a same-level comparison commodity, screens out all commodities which are the same as the class of the secondary intention commodity from the primary intention commodity as a low-level comparison commodity, and records the number of the same-level comparison commodity and the number of the low-level comparison commodities; the product selected from the second level intention products includes at least 1 product, and the number of the lower level control products may be any one of 0, 1 or more.
If the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is greater than 1, executing a step two;
if the number of the same-level comparison commodities and the number of the lower-level comparison commodities are equal to 1, executing a third step;
if the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is 0, executing a fourth step;
if the number of the same-level comparison commodities and the number of the low-level comparison commodities are both larger than 1, executing a fifth step;
if the number of the same-level comparison commodities is more than 1 and the number of the lower-level comparison commodities is equal to 1, executing a step six;
the first-level intention merchandise corresponds to the function of the merchandise during browsing, and the second-level intention merchandise corresponds to the merchandise function having a higher purchase intention, so that the function reserved in the whole process of browsing to prepare for purchasing can be regarded as the most important function of the user, while the function discarded during preparing for purchasing can be regarded as the function of no intention during browsing, and the function added during preparing for purchasing can be regarded as the preferred function because of the added functions during browsing. Based on the principle, different processing methods can be selected according to the specific quantity of the same-level comparison commodities and the lower-level comparison commodities.
If the number of the same-level comparison commodities is more than 1 and the number of the low-level comparison commodities is 0, executing a seventh step;
step two, the keyword extraction module records the function set of the secondary intention commodity as C2The intersection of the functional sets of all low-level control commodities is denoted as JdThe keyword extraction module records the union of the function sets of all the low-level comparison commodities as BdThe keyword extraction module is according to the formula Fb ═ C2∩JdCalculating necessary function set Fb, and then according to the formula Fw ═ Bd-C2That is, the functions included in the first-level intention commodity and not included in the second-level intention commodity are calculated to obtain the irrelevant function set Fw, and finally, the formula Fy is equal to C2-BdCalculating a preferred function set Fy; i.e., functions that are contained in the secondary-intent commodity but not in the primary-intent commodity.
Thirdly, the keyword extraction module records the function set of the secondary intention commodity as C2The function set of the first-level intention commodity is marked as C1The keyword extraction module is according to the formula Fb ═ C1∩C2Calculating the necessary function set Fb, and then according to the formula Fw ═ C1-C2Calculating the irrelevant function set Fw, and finally, according to the formula Fy ═ C2-C1Calculating a preferred function set Fy;
step four, the keyword extraction module takes the empty set as a necessary function set Fb, an irrelevant function set Fw and an optimal function set Fy; the results are not output because the data is not sufficient for analysis.
Step five, the keyword extraction module records the intersection of the function sets of all the low-level comparison commodities as JdThe intersection of the function sets of all the same level comparison commodities is recorded as Jt(ii) a The union of the function sets of all the low-level control commodities is denoted as BdThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module is according to the formula Fb ═ Jt∩JdCalculating a necessary function set Fb; then according to the formula Fw ═ Bd-BtCalculating an irrelevant function set Fw, and finally calculating an optimal function set Fy according to a formula Fy-Bd;
step six, the keyword extraction module records the intersection of the function sets of all the same-level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe function set of the first-level intention commodity is marked as C1The keyword extraction module extracts JtAs the necessary function set Fb, according to the formula Fw ═ C1-BtCalculating the irrelevant function set Fw, and finally, according to the formula Fy being Bt-C1Calculating a preferred function set Fy;
step seven, the keyword extraction module records the intersection of the function sets of all the same level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module extracts JtAs the necessary function set Fb, the empty set is used as the irrelevant function set Fw, and finally, B is obtained according to the formula Fyt-JtCalculating a preferred function set Fy;
and step eight, taking the corresponding functions in the necessary function set Fb, the irrelevant function set Fw and the preferred function set Fy obtained in any one of the steps two to seven as the necessary function key, the irrelevant function key and the preferred function key of the customer for the product.
The specific working process of this embodiment is as follows:
1) acquiring browsing records of a client, simplifying the browsing records according to the time length of single browsing and the browsing times, and determining commodities which the client intends to purchase;
2) the method comprises the following steps of dividing commodities which are intentionally purchased by a customer into two intention levels according to the purchasing habit of the customer, wherein the higher the level is, the higher the purchasing probability is;
3) by matching the functions of the purchased commodities intentionally by the customer with different intention levels, the functions that the customer pays most attention to the commodities of one category and the functions of the commodities that the customer does not pay attention to are determined, and the function of evaluating the commodities is further improved.
The foregoing is merely exemplary and illustrative of the present invention and various modifications, additions and substitutions may be made by those skilled in the art to the specific embodiments described without departing from the scope of the invention as defined in the following claims.
Claims (2)
1. An intention customer analysis system based on big data comprises a client end in communication connection with the system, and is characterized in that the client end is used for generating a browsing record containing a commodity name and browsing duration of a commodity page after a customer clicks a commodity link, and the system further comprises a commodity classification module, a customer behavior acquisition module, a commodity intention grading module, a keyword extraction module, a commodity function extraction module and a storage module;
the data stored by the storage module comprises commodity classification data, a commodity function database of each commodity, browsing records of users, commodity lists in shopping carts and finished order information;
the commodity classification module is used for classifying all commodities according to a commodity classification table and dividing the commodities according to the same class;
the client behavior acquisition module is used for acquiring browsing records, commodity lists and order information of a client from a client;
the commodity function extraction module acquires the functions of commodities from the commodity function databases of various categories and generates a function set of the commodities;
after analyzing the browsing records of the client and the shopping cart list, the commodity intention grading module divides the browsed commodities into a first-level intention and a second-level intention;
the commodity intention grading module specifically grades commodities according to the following steps:
s1, the commodity intention grading module acquires the commodity name and the browsing duration of the commodity page in each browsing record of the client, and deletes the browsing record of which the browsing duration of the commodity page is less than 3 seconds; then obtaining the number of browsing records with the same commodity name as the browsing times, finally deleting the browsing records with the browsing times smaller than 2, and obtaining the commodity names in the rest browsing records as the first-level intention commodities of the client;
s2, the commodity intention grading module calculates the browsing times of the corresponding first-level intention commodity from the rest browsing records as real times, compares the real times with a preset contrast value A, if the real times are larger than the contrast value A, the first-level intention commodity is upgraded to a second-level intention commodity, and meanwhile, the commodity in the commodity list in the shopping cart is also used as the second-level intention commodity;
the keyword extraction module is used for analyzing and comparing function sets of commodities with different intention grades after the commodity intention grading module grades, and screening necessary function keywords, irrelevant function keywords and preferred function keywords of the customers for the functions of the commodities of the class;
the specific steps of the keyword extraction module for acquiring the necessary functional keywords, irrelevant functional keywords and preferred functional keywords of a customer for a commodity of one category are as follows:
step one, a keyword extraction module randomly selects a class corresponding to a secondary intention commodity, screens out commodities of the class from all the secondary intention commodities as a same-level comparison commodity, screens out all commodities which are the same as the class of the secondary intention commodity from the primary intention commodity as a low-level comparison commodity, and records the number of the same-level comparison commodity and the number of the low-level comparison commodities;
if the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is greater than 1, executing a step two;
if the number of the same-level comparison commodities and the number of the lower-level comparison commodities are equal to 1, executing a third step;
if the number of the same-level comparison commodities is equal to 1 and the number of the low-level comparison commodities is 0, executing a fourth step;
if the number of the same-level comparison commodities and the number of the low-level comparison commodities are both larger than 1, executing a fifth step;
if the number of the same-level comparison commodities is more than 1 and the number of the lower-level comparison commodities is equal to 1, executing a step six;
if the number of the same-level comparison commodities is more than 1 and the number of the low-level comparison commodities is 0, executing a seventh step;
step two, the keyword extraction module records the function set of the secondary intention commodity as C2The intersection of the functional sets of all low-level control commodities is denoted as JdThe keyword extraction module is according to the formula Fb ═ C2∩JdCalculating necessary function set Fb, and recording the union of all low-level comparison commodity function sets as B by the keyword extraction moduledThen according to the formula Fw ═ Bd-C2Calculating the irrelevant function set Fw, and finally, according to the formula Fy ═ C2-BdCalculating a preferred function set Fy;
thirdly, the keyword extraction module records the function set of the secondary intention commodity as C2The function set of the first-level intention commodity is marked as C1The keyword extraction module is according to the formula Fb ═ C1∩C2Calculating the necessary function set Fb, and then according to the formula Fw ═ C1-C2Calculating the irrelevant function set Fw, and finally, according to the formula Fy ═ C2-C1Calculating a preferred function set Fy;
step four, the keyword extraction module takes the empty set as a necessary function set Fb, an irrelevant function set Fw and an optimal function set Fy;
step five, the keyword extraction module records the intersection of the function sets of all the low-level comparison commodities as JdThe intersection of the function sets of all the same level comparison commodities is recorded as Jt(ii) a The union of the function sets of all the low-level control commodities is denoted as BdThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module is according to the formula Fb ═ Jt∩JdCalculating a necessary function set Fb; then according to the formula Fw ═ Bd-BtCalculating an irrelevant function set Fw, and finally calculating an optimal function set Fy according to a formula Fy-Bd;
step six, the keyword extraction module records the intersection of the function sets of all the same-level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe function set of the first-level intention commodity is marked as C1The keyword extraction module extracts JtAs the necessary function set Fb, according to the formula Fw ═ C1-BtCalculating the irrelevant function set Fw, and finally, according to the formula Fy being Bt-C1Calculating a preferred function set Fy;
step seven, the keyword extraction module records the intersection of the function sets of all the same level comparison commodities as JtThe function set of all the same level comparison commodities is merged as BtThe keyword extraction module extracts JtAs the necessary function set Fb, the empty set is used as the irrelevant function set Fw, and finally, B is obtained according to the formula Fyt-JtCalculating a preferred function set Fy;
and step eight, correspondingly generating the necessary function keywords, the irrelevant function keywords and the preferred function keywords of the customer for the product class by corresponding functions in the necessary function set Fb, the irrelevant function set Fw and the preferred function set Fy which are obtained in any one of the steps two to seven.
2. The big-data based intention customer analysis system of claim 1, wherein the numerical value of the control value a is obtained by:
screening purchased commodities from the completed order information of the client to be used as reference commodities, screening browsing times corresponding to the reference commodities and the number of the reference commodities from browsing records of the client, calculating average browsing times, and taking the average browsing times as a comparison value A, wherein the calculation formula of the average browsing times is as follows:
T＝(t1+t2+…+tn)/n；
where T is the average number of views, (T1+ T2+ … + tn) is the sum of the number of views for each reference item, n is the number of reference items.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910407588.5A CN110232589B (en) | 2019-05-16 | 2019-05-16 | Intention customer analysis system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910407588.5A CN110232589B (en) | 2019-05-16 | 2019-05-16 | Intention customer analysis system based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110232589A CN110232589A (en) | 2019-09-13 |
| CN110232589B true CN110232589B (en) | 2020-09-15 |
Family
ID=67861359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910407588.5A Active CN110232589B (en) | 2019-05-16 | 2019-05-16 | Intention customer analysis system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110232589B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110716979A (en) * | 2019-10-18 | 2020-01-21 | 重庆锐云科技有限公司 | House buying intention client mining method, device and server |
| CN116756427A (en) * | 2023-06-27 | 2023-09-15 | 鼎翰文化股份有限公司 | Travel information pushing system based on big data |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4596043B2 (en) * | 2008-06-03 | 2010-12-08 | ソニー株式会社 | Information processing apparatus, information processing method, and program |
| CN102609854A (en) * | 2011-01-25 | 2012-07-25 | 青岛理工大学 | Client partitioning method and device based on unified similarity calculation |
| CN104331816B (en) * | 2014-10-28 | 2017-10-20 | 常州大学 | Knowledge based learns big data user's purchase intention Forecasting Methodology with secret protection |
| CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
| CN109559156A (en) * | 2018-10-30 | 2019-04-02 | 广东原昇信息科技有限公司 | Client's intention based on client properties and marketing data has monitoring forecast method |
-
2019
- 2019-05-16 CN CN201910407588.5A patent/CN110232589B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110232589A (en) | 2019-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109685631B (en) | Personalized recommendation method based on big data user behavior analysis | |
| US7299194B1 (en) | Method and system for researching sales effects of advertising using association analysis | |
| JP5368665B2 (en) | Expert database forwarded back to link weighted association rules | |
| CN106708821A (en) | User personalized shopping behavior-based commodity recommendation method | |
| JP6325745B2 (en) | Information processing apparatus, information processing method, and information processing program | |
| US20030055707A1 (en) | Method and system for integrating spatial analysis and data mining analysis to ascertain favorable positioning of products in a retail environment | |
| CN105894332A (en) | Commodity recommendation method, device and system based on user behavior analysis | |
| CN107004203A (en) | Information processing system and method | |
| CN104679771A (en) | Individual data searching method and device | |
| CN111738805B (en) | Behavior log-based search recommendation model generation method, device and storage medium | |
| JP5824602B1 (en) | Information processing apparatus, information processing method, and information processing program | |
| US20120330807A1 (en) | Systems and methods for consumer price index determination using panel-based and point-of-sale market research data | |
| CN110232589B (en) | Intention customer analysis system based on big data | |
| CN110517103A (en) | A kind of Method of Commodity Recommendation and system based on user's collection | |
| CN104537553A (en) | Application of repeated negative sequence pattern in customer purchase behavior analysis | |
| CN113837842A (en) | Commodity recommendation method and equipment based on user behavior data | |
| US8577754B1 (en) | Identifying low utility item-to-item association mappings | |
| CN110765346A (en) | User intention mining method, device and equipment | |
| Srivastava et al. | Improved market basket analysis with utility mining | |
| CN111553762A (en) | Method, system and terminal equipment for improving search quality | |
| CN115860865A (en) | Commodity combination construction method and device, equipment, medium and product thereof | |
| KR102429104B1 (en) | Product catalog automatic classification system based on artificial intelligence | |
| CN112529666A (en) | Commodity recommendation method, commodity recommendation system, computer equipment and storage medium | |
| CN113159828A (en) | Promotion scheme recommendation method and device and computer readable storage medium | |
| Somya et al. | A novel approach to collect and analyze market customer behavior data on online shop |
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 | ||
| CB02 | Change of applicant information |
Address after: 310000 1-206, 206M, 5g Innovation Park, 1818-1 Wenyi West Road, Yuhang District, Hangzhou City, Zhejiang Province Applicant after: ZHEJIANG HUAKUN DAOWEI DATA TECHNOLOGY Co.,Ltd. Address before: 310016 Room 2404, Building A, Hualian Times Building, Jianggan District, Hangzhou City, Zhejiang Province Applicant before: ZHEJIANG HUAKUN DAOWEI DATA TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |