CN117217865A - Personalized recommendation system based on big data analysis - Google Patents
Personalized recommendation system based on big data analysis Download PDFInfo
- Publication number
- CN117217865A CN117217865A CN202311171286.5A CN202311171286A CN117217865A CN 117217865 A CN117217865 A CN 117217865A CN 202311171286 A CN202311171286 A CN 202311171286A CN 117217865 A CN117217865 A CN 117217865A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- unit
- data
- price
- 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.)
- Granted
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims description 36
- 238000007781 pre-processing Methods 0.000 claims description 27
- 238000012795 verification Methods 0.000 claims description 20
- 238000010276 construction Methods 0.000 claims description 19
- 238000000034 method Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 9
- 238000012163 sequencing technique Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a personalized recommendation system based on big data analysis, in particular to the technical field of big data analysis, and the invention obtains the attention index and commodity type demand index of a user for different types of commodities by carrying out a series of processing on the historical shopping behavior data and commodity attribute information of the user, sets a commodity intelligent recommendation module to calculate a commodity purchase difficulty coefficient, and further calculates a commodity comprehensive purchase possibility index.
Description
Technical Field
The invention relates to the technical field of big data analysis, in particular to a personalized recommendation system based on big data analysis.
Background
At present, the big data technology is widely applied to a plurality of industries, in particular to a personalized recommendation service system brought by big data, and is popularized and achieved in the electronic commerce industry.
The conventional personalized recommendation system based on big data is used for calling shopping records of users, extracting commodity types, brands and price information purchased by the users, extracting commodities conforming to characteristic information from a commodity library and sending the commodities to a commodity display interface, and setting a price and sales ordering mode, so that the users can select commodity arrangement display schemes by themselves, and personalized recommendation and real-time update of commodities for the users are realized.
However, the above system still has some problems: the real-time update of commodity recommendation can be realized by carrying out commodity recommendation based on the user shopping record, but the real-time shopping record of the user has contingency, and the real demand and commodity attention characteristics of the user cannot be determined by specific data, so that the accuracy of commodity recommendation prediction needs to be improved.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, an embodiment of the present invention provides a personalized recommendation system based on big data analysis, so as to solve the problems set forth in the above-mentioned background art.
In order to achieve the above purpose, the present invention provides the following technical solutions: a personalized recommendation system based on big data analysis, comprising:
and the user data acquisition module is used for: collecting personal information of a user, historical shopping behavior data and commodity attribute information corresponding to the historical shopping behavior data of the user, and respectively sending the personal information, the historical shopping behavior data and commodity attribute information to a user shopping feature portrait construction module and a shopping associated data preprocessing module;
and a user fund information calling module: after passing the user authority verification, acquiring the accumulated free-standing-use free mobile fund amount under the user name and sending the free mobile fund amount to the commodity intelligent recommendation module;
shopping associated data preprocessing module: preprocessing historical shopping behavior data and commodity attribute information of a user, and respectively calculating the browsing rate, the searching rate, the collecting rate, the purchasing rate and the purchasing rate of the same type of commodity and a commodity preference price interval;
shopping associated data processing module: calculating attention indexes of users for different types of commodities based on the browsing rate, the searching rate, the collection rate and the purchasing rate, and calculating commodity type demand indexes based on the purchasing rates of different commodity types;
the user shopping feature portrait construction module: constructing shopping feature portraits of users based on personal information of the users, commodity price preference intervals, different types of commodity attention indexes and commodity type demand indexes;
and the commodity intelligent recommendation module: carrying out commodity recommendation based on shopping feature images and fund information of the user and intelligently sequencing recommended commodities;
and a recommendation information feedback module: feeding back actual purchase conditions of the user on the system recommended goods and the non-system recommended goods;
the commodity recommendation matching degree calculating module is used for: calculating commodity recommendation matching degree based on commodity recommendation information and recommendation feedback information;
database: for storing all module data in the system.
Preferably, the user data acquisition module comprises a personal information acquisition unit, a historical shopping behavior data acquisition unit, a commodity attribute information extraction unit and an information output unit, wherein the personal information acquisition unit acquires the age, sex and residence geographic position of a user; the historical shopping behavior data acquisition unit acquires commodity browsing residence time, commodity searching records, commodity collection records, commodity purchasing records and commodity purchasing records in a fixed historical time period of a user; the commodity attribute information extraction unit extracts commodity price and category corresponding to user behavior data; the information output unit sends the collected user personal information to the user shopping feature portrait construction module, and sends the collected user historical shopping behavior data and commodity attribute information corresponding to the user historical shopping behavior data to the shopping associated data preprocessing module.
Preferably, the user fund information retrieving module comprises a right verifying unit, a user fund information retrieving unit, a mobile fund information obtaining unit and a data output unit, wherein the right verifying unit confirms whether the user is self-operation before obtaining the user personal fund information, and the identity verification comprises three steps, namely login password verification, face recognition verification and fingerprint verification; the user fund information calling unit calls all account funds under the user name after passing the verification information of the three steps; the mobile fund information acquisition unit deducts all fixed funds and daily necessary expenditure under the user name, and displays the mobile fund amount which can be arbitrarily controlled; and the data output unit sends the mobile fund amount of the current user to the commodity intelligent recommendation module.
Preferably, the shopping associated data preprocessing module comprises an associated commodity classification summarizing unit, a user behavior data processing unit, a commodity data calculating unit, a commodity price interval determining unit and a data output unit, and the specific preprocessing process is as follows:
the associated commodity classification summarization unit: summarizing commodities associated with the user according to commodity types, and marking corresponding commodity type numbers as A i ;
User behavior data processing unit: calculating a browsing rate t based on browsing, searching, collecting, purchasing and purchasing data of different types of commodities by a user in a historical time period x Search Rate b x Collection rate c x Purchasing rate d x And purchase rate e x The specific calculation formula is as follows: wherein t is ci 、b ci 、c ai 、d ci 、e ci Respectively browsing residence time, searching times, collecting times, purchasing times and purchasing amount of each type of commodity, t ce 、b ce 、c ae 、d ce 、e ce The total residence time, the total searching times, the total collecting times, the total purchasing times and the total purchasing amount of the commodities which are browsed in the historical time period are respectively;
commodity preference price interval determining unit: counting the price p of the same kind of commodity associated with the user behavior data in the historical time period i Calculating average value p of similar commodities after eliminating abnormal values ai The specific calculation formula is as follows:calculating price standard deviation sigma pi The specific calculation formula is as follows: />The corresponding price interval is U0, p ai +σ pi ];
A data output unit: and outputting the processed user behavior data and commodity data to a shopping associated data processing module, and outputting the commodity preference price interval of the same type to a user shopping feature portrait construction module.
Preferably, the shopping associated data processing module comprises a data receiving unit, a commodity type attention index calculating unit, a commodity type demand index calculating unit and a data output unit, and the specific data processing process is as follows:
a data receiving unit: receiving data sent by a user data preprocessing module;
commodity type attention index calculating unit: based on browsing rate t x Search Rate b x Collection rate c x Purchasing rate d x Calculating attention index alpha of user for different types of commodities c The specific calculation formula is as follows:a 1 、a 2 、a 3 the parameters are respectively a purchase rate constant coefficient, collection rate experience index and search rate adjustment parameter 1 >a 2 >a 3 >0;
Commodity type demand index calculation unit: purchase rate e based on different commodity types x Calculating commodity type demand index alpha e The specific calculation formula is as follows: alpha e =b 1 *[ln(e x +1)] 2 +b 2 *e x +b t ,b 1 、b 2 Is a constant coefficient, b t To adjust parameters;
A data output unit: and sending the calculated commodity type attention index and commodity type demand index to a user shopping feature portrait construction module.
Preferably, the commodity intelligent recommendation module comprises a data receiving unit, a characteristic information extracting unit, a recommended commodity price judging unit, a recommended commodity price correcting unit, a recommended commodity price condition screening unit, a recommended commodity purchase difficulty coefficient calculating unit, a commodity comprehensive purchase possibility predicting unit, an intelligent sorting unit and a commodity recommendation information output unit, wherein the data receiving unit is used for receiving shopping characteristic images and fund information of a user; the feature information extraction unit is used for extracting a residence geographic position, a commodity price preference interval, a commodity type attention index and a commodity type demand index in the user feature portrait; the recommended commodity price judging unit judges the maximum value pbi of the commodity price of different types preferred by the user and the current flowing fund amount p of the user c Comparing if p bi >p c If the price of the commodity preferred by the user exceeds the consumption level of the user, the upper limit of the price of the commodity recommended by the user is required to be reset, if p bi ≤p c Judging that the price of the commodity preferred by the user accords with the consumption level of the user, wherein the original commodity price preference interval is a recommended commodity price interval; the recommended commodity price correction unit resets the recommended commodity price interval when the user prefers commodity price to exceed the user consumption level, and the new recommended commodity price interval is U, [0, p ] c ]The method comprises the steps of carrying out a first treatment on the surface of the The recommended commodity price condition screening unit screens commodities based on recommended price intervals of different types of commodities, and reserves commodity contents meeting recommended price conditions; the commodity price p meeting the recommended price condition is calculated by the recommended commodity purchase difficulty coefficient calculation unit d With the current amount p of funds to be deposited by the user c Data comparison is carried out to calculate the difficulty coefficient gamma of commodity purchase recommendation a The specific calculation formula is as follows:a 0 to recommend the purchase difficulty factor adjustment factor for the merchandise,a 0 >1, a step of; the comprehensive commodity purchase possibility prediction unit is based on the screened commodity type attention index alpha c Commodity type demand index alpha e Coefficient of difficulty in purchasing recommended commodity gamma a Calculating a comprehensive purchase likelihood index W of a commodity a The specific calculation formula is as follows:x 1 、x 2 to exponentially adjust the parameters, phi a Is a commodity attribute influence factor; the intelligent sorting unit sorts the commodity comprehensive purchase probability index values from large to small based on the calculated commodity comprehensive purchase probability index values, and the specific rank is recorded as L j If the calculated comprehensive purchase probability index values of the commodities are the same, sorting the commodities according to the distances between commodity delivery addresses and the geographical positions of the residence places of the users from near to far; and the commodity recommendation information output unit sends the commodities subjected to intelligent sequencing to a commodity recommendation display page.
Preferably, the recommendation information feedback module comprises a recommendation commodity purchase data feedback unit, a non-recommendation commodity purchase data feedback unit and a data output unit, wherein the recommendation commodity purchase data feedback unit feeds back the high-to-low ranking L of purchased recommendation commodities in intelligent sorting i And recommended commodity purchase quantity n a The method comprises the steps of carrying out a first treatment on the surface of the The non-recommended commodity purchase data feedback unit feeds back the non-recommended commodity purchase quantity n b The method comprises the steps of carrying out a first treatment on the surface of the And the data output unit sends the fed-back data to the commodity recommendation matching degree calculation module.
Preferably, the commodity recommendation matching degree calculating module calculates the recommendation matching degree beta based on commodity recommendation information and recommendation feedback information e The specific calculation formula of (2) is as follows:L j the desired commodity rank is ranked for intelligent ordering.
The invention has the technical effects and advantages that:
1. the invention sets the user data acquisition module to acquire commodity browsing stay time, commodity searching record, commodity collection record, commodity purchasing record, commodity price and category information in a fixed historical time period of a user, sets the shopping associated data preprocessing module to calculate browsing rate, searching rate, collection rate and purchasing rate based on browsing, searching, collecting and purchasing data of different types of commodities in the historical time period, sets the shopping associated data processing module to calculate attention index of the user to different types of commodities based on the browsing rate, searching rate, collecting rate and purchasing rate, calculates commodity type demand index based on the purchasing rate of different commodity types, sets the commodity intelligent recommendation module to calculate commodity purchasing difficulty coefficient based on commodity type attention index and commodity purchasing difficulty coefficient, and compared with the prior art, the acquired data range of the invention is wider, the calculated commodity type attention index shows attention index of the user to different types of commodities, the actual purchasing demand of the user to different types of commodities is more accurate, and the calculated commodity type attention index accords with the actual purchasing requirement of the user to the commodity.
2. The invention sets the user fund information invoking module to acquire the accumulated free-use free-form dominating mobile fund amount under the user name after passing the user authority verification, sets the shopping associated data preprocessing module to determine the commodity preference price interval based on the same commodity price associated with the user behavior data in the historical time period, sets the commodity intelligent recommending module to compare the maximum value of the commodity price of different types of the user preference with the current mobile fund amount of the user, judges that the commodity price of the user preference exceeds the user consumption level if the maximum value of the commodity price of different types of the user preference is greater than the current mobile fund amount of the user, and resets the upper limit of the commodity price of the recommendation, wherein the new upper limit of the commodity price of the recommendation is the current mobile fund amount of the user, otherwise, judges that the commodity price of the user preference accords with the user consumption level, the original commodity price preference interval is the commodity price of the recommendation interval, screens the recommended commodities and calculates the purchase difficulty coefficient of each commodity on the basis, ensures that the recommended commodities are all within the consumption limit of the user, more fits the economic level of the user, and realizes the purpose of targeted recommendation.
Drawings
Fig. 1 is a block diagram of a system architecture of the present invention.
FIG. 2 is a flow chart of the system operation of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment as shown in fig. 1 provides a personalized recommendation system based on big data analysis, which comprises a user data acquisition module, a user fund information retrieval module, a shopping associated data preprocessing module, a shopping associated data processing module, a user shopping feature image construction module, a commodity intelligent recommendation module, a recommendation information feedback module, a commodity recommendation matching degree calculation module and a database, wherein the user data acquisition module is connected with the shopping associated data preprocessing module and the user shopping feature image construction module, the user fund information retrieval module is connected with the commodity intelligent recommendation module, the shopping associated data preprocessing module is connected with the associated data processing module and the user shopping feature image construction module, and the shopping associated data processing module, the user shopping feature image construction module, the commodity intelligent recommendation module, the recommendation information feedback module and the commodity recommendation matching degree calculation module are sequentially connected with the database.
The user data acquisition module acquires personal information of a user, historical shopping behavior data and commodity attribute information corresponding to the historical shopping behavior data of the user and sends the personal information, the historical shopping behavior data and commodity attribute information to the user shopping feature portrait construction module and the shopping associated data preprocessing module respectively.
Further, the user data acquisition module comprises a personal information acquisition unit, a historical shopping behavior data acquisition unit, a commodity attribute information extraction unit and an information output unit, wherein the personal information acquisition unit acquires the age, sex and residence geographic position of a user; the historical shopping behavior data acquisition unit acquires commodity browsing residence time, commodity searching records, commodity collection records, commodity purchasing records and commodity purchasing records in a fixed historical time period of a user; the commodity attribute information extraction unit extracts commodity price and category corresponding to user behavior data; the information output unit sends the collected user personal information to the user shopping feature portrait construction module, and sends the collected user historical shopping behavior data and commodity attribute information corresponding to the user historical shopping behavior data to the shopping associated data preprocessing module.
And the user fund information calling module acquires the free-standing-use free mobile fund amount accumulated under the user name after passing the user authority verification and sends the free-standing-use free mobile fund amount to the commodity intelligent recommending module.
The user fund information retrieving module comprises a right verifying unit, a user fund information retrieving unit, a mobile fund information obtaining unit and a data output unit, wherein the right verifying unit confirms whether the user is operated by the user before obtaining the user personal fund information, and the identity verification comprises three steps, namely login password verification, face recognition verification and fingerprint verification; the user fund information calling unit calls all account funds under the user name after passing the verification information of the three steps; the mobile fund information acquisition unit deducts all fixed funds and daily necessary expenditure under the user name, and displays the mobile fund amount which can be arbitrarily controlled; and the data output unit sends the mobile fund amount of the current user to the commodity intelligent recommendation module.
The shopping associated data preprocessing module is used for preprocessing historical shopping behavior data and commodity attribute information of a user and respectively calculating the browsing rate, the searching rate, the collection rate, the purchasing rate and the commodity preference price interval of the same type of commodities.
Further, the shopping associated data preprocessing module comprises an associated commodity classification summarizing unit, a user behavior data processing unit, a commodity data calculating unit, a commodity price interval determining unit and a data output unit, and the specific preprocessing process is as follows:
the associated commodity classification summarization unit: summarizing commodities associated with the user according to commodity types, and marking corresponding commodity type numbers as A i ;
User behavior data processing unit: calculating a browsing rate t based on browsing, searching, collecting, purchasing and purchasing data of different types of commodities by a user in a historical time period x Search Rate b x Collection rate c x Purchasing rate d x And purchase rate e x The specific calculation formula is as follows: wherein t is ci 、b ci 、c ai 、d ci 、e ci Respectively browsing residence time, searching times, collecting times, purchasing times and purchasing amount of each type of commodity, t ce 、b ce 、c ae 、d ce 、e ce The total residence time, the total searching times, the total collecting times, the total purchasing times and the total purchasing amount of the commodities which are browsed in the historical time period are respectively;
commodity preference price interval determining unit: counting the price p of the same kind of commodity associated with the user behavior data in the historical time period i Calculating average value p of similar commodities after eliminating abnormal values ai The specific calculation formula is as follows:calculating price standard deviation sigma pi The specific calculation formula is as follows: />The corresponding price interval is U0, p ai +σ pi ];
A data output unit: and outputting the processed user behavior data and commodity data to a shopping associated data processing module, and outputting the commodity preference price interval of the same type to a user shopping feature portrait construction module.
The shopping associated data processing module calculates attention indexes of users for different types of commodities based on browsing rate, searching rate, collection rate and purchasing rate, and calculates commodity type demand indexes based on purchasing rates of different commodity types.
Further, the shopping associated data processing module comprises a data receiving unit, a commodity type attention index calculating unit, a commodity type demand index calculating unit and a data output unit, and the specific data processing process is as follows:
a data receiving unit: receiving data sent by a user data preprocessing module;
commodity type attention index calculating unit: based on browsing rate t x Search Rate b x Collection rate c x Purchasing rate d x Calculating attention index alpha of user for different types of commodities c The specific calculation formula is as follows:a 1 、a 2 、a 3 the parameters are respectively a purchase rate constant coefficient, collection rate experience index and search rate adjustment parameter 1 >a 2 >a 3 >0;
Commodity type demand index calculation unit: purchase rate e based on different commodity types x Calculating commodity type demand index alpha e The specific calculation formula is as follows: alpha e =b 1 *[ln(e x +1)] 2 +b 2 *e x +b t ,b 1 、b 2 Is a constant coefficient, b t To adjust parameters;
a data output unit: and sending the calculated commodity type attention index and commodity type demand index to a user shopping feature portrait construction module.
The user shopping feature portrait construction module: and constructing shopping feature portraits of the users based on personal information of the users, commodity price preference intervals, different types of commodity attention indexes and commodity type requirement indexes.
The commodity intelligent recommendation module is as follows: and recommending commodities based on the shopping feature images and the fund information of the user, and intelligently sequencing the recommended commodities.
Further, the commodity intelligent recommendation module comprises a data receiving unit, a characteristic information extracting unit, a recommended commodity price judging unit, a recommended commodity price correcting unit, a recommended commodity price condition screening unit, a recommended commodity purchase difficulty coefficient calculating unit, a commodity comprehensive purchase possibility predicting unit, an intelligent sorting unit and a commodity recommendation information output unit, wherein the data receiving unit is used for receiving shopping characteristic images and fund information of a user; the feature information extraction unit is used for extracting a residence geographic position, a commodity price preference interval, a commodity type attention index and a commodity type demand index in the user feature portrait; the recommended commodity price judging unit judges the maximum value pbi of the commodity price of different types preferred by the user and the current flowing fund amount p of the user c Comparing if p bi >p c If the price of the commodity preferred by the user exceeds the consumption level of the user, the upper limit of the price of the commodity recommended by the user is required to be reset, if p bi ≤p c Judging that the price of the commodity preferred by the user accords with the consumption level of the user, wherein the original commodity price preference interval is a recommended commodity price interval; the recommended commodity price correction unit resets the recommended commodity price interval when determining that the user-preferred commodity price exceeds the user consumption level, and the new recommended commodity price interval is U' [0, p ] c ]The method comprises the steps of carrying out a first treatment on the surface of the The recommended commodity price condition screening unit screens commodities based on recommended price intervals of different types of commodities, and reserves commodity contents meeting recommended price conditions; the commodity price p meeting the recommended price condition is calculated by the recommended commodity purchase difficulty coefficient calculation unit d Is current with the userAmount of funds on stream p c Data comparison is carried out to calculate the difficulty coefficient gamma of commodity purchase recommendation a The specific calculation formula is as follows:a 0 adjustment factor for difficulty coefficient of commodity recommendation purchase, a 0 >1, a step of; the comprehensive commodity purchase possibility prediction unit is based on the screened commodity type attention index alpha c Commodity type demand index alpha e Coefficient of difficulty in purchasing recommended commodity gamma a Calculating a comprehensive purchase likelihood index W of a commodity a The specific calculation formula is as follows:x 1 、x 2 adjusting parameter x for index 1 >0、x 2 >0,Φ a Is a commodity attribute influence factor; the intelligent sorting unit sorts the commodity comprehensive purchase probability index values from large to small based on the calculated commodity comprehensive purchase probability index values, and the specific rank is recorded as L j If the calculated comprehensive purchase probability index values of the commodities are the same, sorting the commodities according to the distances between commodity delivery addresses and the geographical positions of the residence places of the users from near to far; and the commodity recommendation information output unit sends the commodities subjected to intelligent sequencing to a commodity recommendation display page.
In this embodiment, it is specifically required to explain that the commodity attribute influence factor is related to sales and good scores of recommended commodities, and a specific calculation formula is as follows:wherein m is a For good value of commodity, m b C is the comment quantity of the commodity a 、c b C is a self-compensating coefficient a >0、c b >0,y a Is the total sales of the commodity.
In this embodiment, it is specifically required to explain that, in order to facilitate understanding of the intelligent sorting situation, a set of numbered comprehensive purchase likelihood indexes are provided: f (f) 1 =10、f 2 =9、f 3 =7、f 4 =7、f 5 =5、f 6 =2, commodity conveyance distance corresponding to: f (f) 1 =8km、f 2 =9km、f 3 =5km、f 4 =4km、f 5 =2km、f 6 =7 km, then its corresponding smart rank ranking is: l (L) 1 =f 1 =1、L 2 =f 2 =2、L 3 =f 4 =3、L 4 =f 3 =4、L 5 =f 5 =5、L 6 =f 6 ==6。
And the recommendation information feedback module feeds back actual purchase conditions of the user on the system recommended goods and the non-system recommended goods.
Further, the recommendation information feedback module comprises a recommendation commodity purchase data feedback unit, a non-recommendation commodity purchase data feedback unit and a data output unit, wherein the recommendation commodity purchase data feedback unit feeds back the order ranking L of purchased recommendation commodities in intelligent sorting i And recommended commodity purchase quantity n a The method comprises the steps of carrying out a first treatment on the surface of the The non-recommended commodity purchase data feedback unit feeds back the non-recommended commodity purchase quantity n b The method comprises the steps of carrying out a first treatment on the surface of the And the data output unit sends the fed-back data to the commodity recommendation matching degree calculation module.
The commodity recommendation matching degree calculating module calculates recommendation matching degree based on commodity recommendation information and recommendation feedback information.
Further, the commodity recommendation matching degree calculating module calculates a recommendation matching degree beta based on commodity recommendation information and recommendation feedback information e The specific calculation formula of (2) is as follows:L j the desired commodity rank is ranked for intelligent ordering.
In this embodiment, it is specifically described that, for easy understandingA set of data is now provided for explanation, assuming that the ranking of the actual purchased items in the intelligent ordering is L 3 、L 7 、L 11 、L 5 、L 1 The expected purchase goods should be ranked in intelligent ordering as L 1 、L 2 、L 3 、L 4 、L 5 There is->
In this embodiment, it is specifically required to explain that the adjustment parameters, the empirical indexes, and the constant coefficients used in the formulas in this embodiment are all selected based on the actual requirements, and are not limited to specific numerical values.
The database is used for storing all module data in the system.
The embodiment as shown in fig. 2 provides an operation flow of a personalized recommendation system based on big data analysis, which comprises the following steps:
s1: collecting the age, sex, residence geographic position, commodity browsing residence time, commodity searching record, commodity collection record, commodity purchasing record, commodity price and variety of a user in a fixed historical time period;
s2: acquiring the accumulated free-standing free mobile fund amount under the user name after passing the user authority verification;
s3: preprocessing historical shopping behavior data and commodity attribute information of a user, and respectively calculating the browsing rate, the searching rate, the collecting rate, the purchasing rate and the purchasing rate of the same type of commodity and a commodity preference price interval;
s4: calculating attention indexes of users for different types of commodities based on the browsing rate, the searching rate, the collection rate and the purchasing rate, and calculating commodity type demand indexes based on the purchasing rates of different commodity types;
s5: constructing shopping feature portraits of users based on personal information of the users, commodity price preference intervals, different types of commodity attention indexes and commodity type demand indexes;
s6: carrying out commodity recommendation based on shopping feature images and fund information of the user and intelligently sequencing recommended commodities;
s7: feeding back actual purchase conditions of the user on the system recommended goods and the non-system recommended goods;
s8: and calculating commodity recommendation matching degree based on the commodity recommendation information and the recommendation feedback information.
Finally: the foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (8)
1. The personalized recommendation system based on big data analysis is characterized in that: comprising the following steps:
and the user data acquisition module is used for: collecting personal information of a user, historical shopping behavior data and commodity attribute information corresponding to the historical shopping behavior data of the user, and respectively sending the personal information, the historical shopping behavior data and commodity attribute information to a user shopping feature portrait construction module and a shopping associated data preprocessing module;
and a user fund information calling module: after passing the user authority verification, acquiring the accumulated free-standing-use free mobile fund amount under the user name and sending the free mobile fund amount to the commodity intelligent recommendation module;
shopping associated data preprocessing module: preprocessing historical shopping behavior data and commodity attribute information of a user, and respectively calculating the browsing rate, the searching rate, the collecting rate, the purchasing rate and the purchasing rate of the same type of commodity and a commodity preference price interval;
shopping associated data processing module: calculating attention indexes of users for different types of commodities based on the browsing rate, the searching rate, the collection rate and the purchasing rate, and calculating commodity type demand indexes based on the purchasing rates of different commodity types;
the user shopping feature portrait construction module: constructing shopping feature portraits of users based on personal information of the users, commodity price preference intervals, different types of commodity attention indexes and commodity type demand indexes;
and the commodity intelligent recommendation module: carrying out commodity recommendation based on shopping feature images and fund information of the user and intelligently sequencing recommended commodities;
and a recommendation information feedback module: feeding back actual purchase conditions of the user on the system recommended goods and the non-system recommended goods;
the commodity recommendation matching degree calculating module is used for: and calculating commodity recommendation matching degree based on the commodity recommendation information and the recommendation feedback information.
2. The personalized recommendation system based on big data analysis of claim 1, wherein: the user data acquisition module comprises a personal information acquisition unit, a historical shopping behavior data acquisition unit, a commodity attribute information extraction unit and an information output unit, wherein the personal information acquisition unit acquires the age, sex and residence geographic position of a user; the historical shopping behavior data acquisition unit acquires commodity browsing residence time, commodity searching records, commodity collection records, commodity purchasing records and commodity purchasing records in a fixed historical time period of a user; the commodity attribute information extraction unit extracts commodity price and category corresponding to user behavior data; the information output unit sends the collected user personal information to the user shopping feature portrait construction module, and sends the collected user historical shopping behavior data and commodity attribute information corresponding to the user historical shopping behavior data to the shopping associated data preprocessing module.
3. The personalized recommendation system based on big data analysis of claim 1, wherein: the user fund information retrieving module comprises an authority verifying unit, a user fund information retrieving unit, a mobile fund information obtaining unit and a data output unit, wherein the authority verifying unit confirms whether the user is personally operated before obtaining user personal fund information, and the identity verification comprises three steps, namely login password verification, face recognition verification and fingerprint verification; the user fund information calling unit calls all account funds under the user name after passing the verification information of the three steps; the mobile fund information acquisition unit deducts all fixed funds and daily necessary expenditure under the user name, and displays the mobile fund amount which can be arbitrarily controlled; and the data output unit sends the mobile fund amount of the current user to the commodity intelligent recommendation module.
4. The personalized recommendation system based on big data analysis of claim 1, wherein: the shopping associated data preprocessing module comprises an associated commodity classification summarizing unit, a user behavior data processing unit, a commodity data calculating unit, a commodity price interval determining unit and a data output unit, and the specific preprocessing process is as follows:
the associated commodity classification summarization unit: summarizing commodities associated with the user according to commodity types, and marking corresponding commodity type numbers as A i ;
User behavior data processing unit: calculating a browsing rate t based on browsing, searching, collecting, purchasing and purchasing data of different types of commodities by a user in a historical time period x Search Rate b x Collection rate c x Purchasing rate d x And purchase rate e x The specific calculation formula is as follows: wherein t is ci 、b ci 、c ai 、d ci 、e ci Respectively browsing residence time, searching times, collecting times, purchasing times and purchasing amount of each type of commodity, t ce 、b ce 、c ae 、d ce 、e ce The total residence time, the total searching times, the total collecting times, the total purchasing times and the total purchasing amount of the commodities which are browsed in the historical time period are respectively;
commodity preference price interval determining unit: counting the price p of the same kind of commodity associated with the user behavior data in the historical time period i Calculating average value p of similar commodities after eliminating abnormal values ai The specific calculation formula is as follows:calculating price standard deviation sigma pi The specific calculation formula is as follows: />The corresponding price interval is U0, p ai +σ pi ];
A data output unit: and outputting the processed user behavior data and commodity data to a shopping associated data processing module, and outputting the commodity preference price interval of the same type to a user shopping feature portrait construction module.
5. The personalized recommendation system based on big data analysis of claim 1, wherein: the shopping associated data processing module comprises a data receiving unit, a commodity type attention index calculating unit, a commodity type demand index calculating unit and a data output unit, and the specific data processing process is as follows:
a data receiving unit: receiving data sent by a user data preprocessing module;
commodity type attention index calculating unit: based on browsing rate t x Search Rate b x Collection rate c x Purchasing rate d x Calculating attention index alpha of user for different types of commodities c The specific calculation formula is as follows:a 1 、a 2 、a 3 the parameters are respectively a purchase rate constant coefficient, collection rate experience index and search rate adjustment parameter 1 >a 2 >a 3 >0;
Commodity type demand index calculation unit: purchase rate e based on different commodity types x Calculating commodity type demand index alpha e The specific calculation formula is as follows: alpha e =b 1 *[ln(e x +1)] 2 +b 2 *e x +b t ,b 1 、b 2 Is a constant coefficient, b t To adjust the ginsengA number;
a data output unit: and sending the calculated commodity type attention index and commodity type demand index to a user shopping feature portrait construction module.
6. The personalized recommendation system based on big data analysis of claim 1, wherein: the commodity intelligent recommendation module comprises a data receiving unit, a characteristic information extracting unit, a recommended commodity price judging unit, a recommended commodity price correcting unit, a recommended commodity price condition screening unit, a recommended commodity purchase difficulty coefficient calculating unit, a commodity comprehensive purchase possibility predicting unit, an intelligent sorting unit and a commodity recommendation information outputting unit, wherein the data receiving unit is used for receiving shopping characteristic portraits and fund information of a user; the feature information extraction unit is used for extracting a residence geographic position, a commodity price preference interval, a commodity type attention index and a commodity type demand index in the user feature portrait; the recommended commodity price judging unit judges the maximum value pbi of the commodity price of different types preferred by the user and the current flowing fund amount p of the user c Comparing if p bi >p c If the price of the commodity preferred by the user exceeds the consumption level of the user, the upper limit of the price of the commodity recommended by the user is required to be reset, if p bi ≤p c Judging that the price of the commodity preferred by the user accords with the consumption level of the user, wherein the original commodity price preference interval is a recommended commodity price interval; the recommended commodity price correction unit resets the recommended commodity price interval when the user prefers commodity price to exceed the user consumption level, and the new recommended commodity price interval is U, [0, p ] c ]The method comprises the steps of carrying out a first treatment on the surface of the The recommended commodity price condition screening unit screens commodities based on recommended price intervals of different types of commodities, and reserves commodity contents meeting recommended price conditions; the commodity price p meeting the recommended price condition is calculated by the recommended commodity purchase difficulty coefficient calculation unit d With the current amount p of funds to be deposited by the user c Data comparison is carried out to calculate the difficulty coefficient gamma of commodity purchase recommendation a The specific calculation formula is as follows:a 0 adjustment factor for difficulty coefficient of commodity recommendation purchase, a 0 >1, a step of; the comprehensive commodity purchase possibility prediction unit is based on the screened commodity type attention index alpha c Commodity type demand index alpha e Coefficient of difficulty in purchasing recommended commodity gamma a Calculating a comprehensive purchase likelihood index W of a commodity a The specific calculation formula is as follows: />x 1 、x 2 To exponentially adjust the parameters, phi a Is a commodity attribute influence factor; the intelligent sorting unit sorts the commodity comprehensive purchase probability index values from large to small based on the calculated commodity comprehensive purchase probability index values, and the specific rank is recorded as L j If the calculated comprehensive purchase probability index values of the commodities are the same, sorting the commodities according to the distances between commodity delivery addresses and the geographical positions of the residence places of the users from near to far; and the commodity recommendation information output unit sends the commodities subjected to intelligent sequencing to a commodity recommendation display page.
7. The personalized recommendation system based on big data analysis of claim 1, wherein: the recommended information feedback module comprises a recommended commodity purchase data feedback unit, a non-recommended commodity purchase data feedback unit and a data output unit, wherein the recommended commodity purchase data feedback unit feeds back the high-to-low ranking L of the purchased recommended commodities in intelligent sorting i And recommended commodity purchase quantity n a The method comprises the steps of carrying out a first treatment on the surface of the The non-recommended commodity purchase data feedback unit feeds back the non-recommended commodity purchase quantity n b The method comprises the steps of carrying out a first treatment on the surface of the And the data output unit sends the fed-back data to the commodity recommendation matching degree calculation module.
8. The personalized recommendation system based on big data analysis of claim 1, wherein: the commodity recommendation matching degree calculating module calculates recommendation matching based on commodity recommendation information and recommendation feedback informationDegree beta e The specific calculation formula of (2) is as follows:L j the desired commodity rank is ranked for intelligent ordering.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311171286.5A CN117217865B (en) | 2023-09-12 | 2023-09-12 | Personalized recommendation system based on big data analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311171286.5A CN117217865B (en) | 2023-09-12 | 2023-09-12 | Personalized recommendation system based on big data analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117217865A true CN117217865A (en) | 2023-12-12 |
CN117217865B CN117217865B (en) | 2024-06-11 |
Family
ID=89048934
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311171286.5A Active CN117217865B (en) | 2023-09-12 | 2023-09-12 | Personalized recommendation system based on big data analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117217865B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117670404A (en) * | 2024-02-02 | 2024-03-08 | 深圳市卖点科技股份有限公司 | Intelligent retail management system and method for intelligent display terminal |
CN117974270A (en) * | 2024-02-02 | 2024-05-03 | 浙江惠利玛产业互联网有限公司 | AI model-based footwear data management method and system |
CN118301210A (en) * | 2024-04-22 | 2024-07-05 | 广东玖诚网络科技有限公司 | Personalized pushing method and system based on big data and deep learning |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090299822A1 (en) * | 2006-11-08 | 2009-12-03 | P C Grocery Ltd. | System and method for optimized shopping transactions |
WO2012151401A1 (en) * | 2011-05-04 | 2012-11-08 | Etsy, Inc. | Item discovery tools and methods for shopping in an electronic commerce environment |
KR20160065429A (en) * | 2014-11-30 | 2016-06-09 | 주식회사 알티웍스 | Hybrid personalized product recommendation method |
CN106485562A (en) * | 2015-09-01 | 2017-03-08 | 苏宁云商集团股份有限公司 | A kind of commodity information recommendation method based on user's history behavior and system |
CN112184399A (en) * | 2020-11-10 | 2021-01-05 | 广州鸿森资本管理有限公司 | Commodity sale recommendation system and method based on Internet |
CN112488807A (en) * | 2020-12-19 | 2021-03-12 | 闫耀伟 | Agricultural product sale recommendation method based on big data |
KR20210066514A (en) * | 2019-11-28 | 2021-06-07 | (주)양유 | Marketing solution system that can analyze and manage the impact of the influencer and Marketing solution method using the same |
CN113034238A (en) * | 2021-04-13 | 2021-06-25 | 武汉畅来电子商务有限公司 | Commodity brand feature extraction and intelligent recommendation management method based on electronic commerce platform transaction |
CN113052651A (en) * | 2021-03-17 | 2021-06-29 | 广东便捷神科技股份有限公司 | Intelligent retail management method and system based on big data |
CN114820140A (en) * | 2022-05-26 | 2022-07-29 | 上海麦开信息科技有限公司 | Furniture full-link digital operation system based on big data analysis |
CN115330476A (en) * | 2022-06-17 | 2022-11-11 | 江苏中创供应链服务有限公司 | Cross-border retail imported commodity intelligent recommendation system based on big data |
US20230099386A1 (en) * | 2020-03-02 | 2023-03-30 | Beijing Wodong Tianjun Information Technology Co., Ltd. | System of recommending item, method of recommending item, computer system, and medium |
-
2023
- 2023-09-12 CN CN202311171286.5A patent/CN117217865B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090299822A1 (en) * | 2006-11-08 | 2009-12-03 | P C Grocery Ltd. | System and method for optimized shopping transactions |
WO2012151401A1 (en) * | 2011-05-04 | 2012-11-08 | Etsy, Inc. | Item discovery tools and methods for shopping in an electronic commerce environment |
KR20160065429A (en) * | 2014-11-30 | 2016-06-09 | 주식회사 알티웍스 | Hybrid personalized product recommendation method |
CN106485562A (en) * | 2015-09-01 | 2017-03-08 | 苏宁云商集团股份有限公司 | A kind of commodity information recommendation method based on user's history behavior and system |
KR20210066514A (en) * | 2019-11-28 | 2021-06-07 | (주)양유 | Marketing solution system that can analyze and manage the impact of the influencer and Marketing solution method using the same |
US20230099386A1 (en) * | 2020-03-02 | 2023-03-30 | Beijing Wodong Tianjun Information Technology Co., Ltd. | System of recommending item, method of recommending item, computer system, and medium |
CN112184399A (en) * | 2020-11-10 | 2021-01-05 | 广州鸿森资本管理有限公司 | Commodity sale recommendation system and method based on Internet |
CN112488807A (en) * | 2020-12-19 | 2021-03-12 | 闫耀伟 | Agricultural product sale recommendation method based on big data |
CN113052651A (en) * | 2021-03-17 | 2021-06-29 | 广东便捷神科技股份有限公司 | Intelligent retail management method and system based on big data |
CN113034238A (en) * | 2021-04-13 | 2021-06-25 | 武汉畅来电子商务有限公司 | Commodity brand feature extraction and intelligent recommendation management method based on electronic commerce platform transaction |
CN114820140A (en) * | 2022-05-26 | 2022-07-29 | 上海麦开信息科技有限公司 | Furniture full-link digital operation system based on big data analysis |
CN115330476A (en) * | 2022-06-17 | 2022-11-11 | 江苏中创供应链服务有限公司 | Cross-border retail imported commodity intelligent recommendation system based on big data |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117670404A (en) * | 2024-02-02 | 2024-03-08 | 深圳市卖点科技股份有限公司 | Intelligent retail management system and method for intelligent display terminal |
CN117974270A (en) * | 2024-02-02 | 2024-05-03 | 浙江惠利玛产业互联网有限公司 | AI model-based footwear data management method and system |
CN117670404B (en) * | 2024-02-02 | 2024-05-28 | 深圳市卖点科技股份有限公司 | Intelligent retail management system and method for intelligent display terminal |
CN118301210A (en) * | 2024-04-22 | 2024-07-05 | 广东玖诚网络科技有限公司 | Personalized pushing method and system based on big data and deep learning |
Also Published As
Publication number | Publication date |
---|---|
CN117217865B (en) | 2024-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117217865B (en) | Personalized recommendation system based on big data analysis | |
US11687961B2 (en) | System, method and computer program product for geo-specific vehicle pricing | |
CN111080398B (en) | Commodity recommendation method, commodity recommendation device, computer equipment and storage medium | |
US8498992B2 (en) | Item selecting apparatus and method, and computer program | |
CN111709812A (en) | E-commerce platform commodity recommendation method and system based on user dynamic classification | |
CN106934498A (en) | The recommendation method and system of hotel's house type in OTA websites | |
US20120130848A1 (en) | Apparatus, Method, And Computer Program For Selecting Items | |
CN108550068A (en) | A kind of individual commodity recommendation method and system based on user behavior analysis | |
EP3399491A1 (en) | Data processing method and device | |
CN104615741B (en) | Cold-start project recommendation method and device based on cloud computing | |
CN110674391B (en) | Product data pushing method and system based on big data and computer equipment | |
JP5831204B2 (en) | Information providing system, information providing method, and program | |
CN110880127B (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
CN116452261B (en) | Advertisement delivery data processing method based on cross-border E-commerce service platform | |
CN113034238B (en) | Commodity brand feature extraction and intelligent recommendation management method based on electronic commerce platform transaction | |
CN112950256A (en) | Method and system for pushing customized advertisement form based on App | |
CN111915409A (en) | Article recommendation method, device and equipment based on article and storage medium | |
CN113052651A (en) | Intelligent retail management method and system based on big data | |
CN116012100A (en) | Electronic commerce platform based on big data analysis | |
CN114493361A (en) | Effectiveness evaluation method and device for commodity recommendation algorithm | |
KR20210003421A (en) | Method and sever for sorting user preferred post | |
CN118313906B (en) | Personalized product recommendation method | |
CN117436936B (en) | Sales prediction and BOM finished product processing system and method | |
KR102696245B1 (en) | System for automated product trading based on artificial intelligence | |
KR102184700B1 (en) | Method for recommending hot websites and server using the same |
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 |