CN115423575B - Internet-based digital analysis management system and method - Google Patents
Internet-based digital analysis management system and method Download PDFInfo
- Publication number
- CN115423575B CN115423575B CN202211366741.2A CN202211366741A CN115423575B CN 115423575 B CN115423575 B CN 115423575B CN 202211366741 A CN202211366741 A CN 202211366741A CN 115423575 B CN115423575 B CN 115423575B
- Authority
- CN
- China
- Prior art keywords
- commodity
- pkj
- style
- time
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000010252 digital analysis Methods 0.000 title claims abstract description 20
- 238000007726 management method Methods 0.000 claims abstract description 28
- 238000012216 screening Methods 0.000 claims abstract description 23
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000007405 data analysis Methods 0.000 claims abstract description 7
- 238000012552 review Methods 0.000 claims description 29
- 238000000605 extraction Methods 0.000 claims description 13
- 230000002349 favourable effect Effects 0.000 claims description 9
- 101001001646 Coturnix japonica Serine/threonine-protein kinase pim-3 Proteins 0.000 claims description 6
- 230000008451 emotion Effects 0.000 claims description 5
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 3
- 230000009471 action Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 102220091149 rs145688699 Human genes 0.000 description 1
- 102220295519 rs779762640 Human genes 0.000 description 1
- 102220086336 rs864622288 Human genes 0.000 description 1
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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of big data, in particular to a digital analysis management system and a digital analysis management method based on the Internet, wherein the system comprises a sales volume prediction module, the sales volume prediction module predicts the sales volume corresponding to each style commodity in the subsequent second unit time t2 based on the current time according to the analysis results in a sales volume data analysis module and a comment analysis module, and the time node corresponding to the current time in the time period is marked as t1. According to the invention, not only can the change condition of the sales volume of the commodity class along with the time node in the time period be obtained by analyzing the historical data, but also the invalid evaluation in the specific evaluation information of the commodity can be screened, so that the interference of the invalid comment on the merchant in the process of analyzing the sales condition of the commodity is avoided, and the sales volume of the commodity in a certain time period can be accurately and effectively predicted by combining the quantized value of the remaining evaluation after screening, thereby further influencing the management of the merchant on the inventory commodity.
Description
Technical Field
The invention relates to the technical field of big data, in particular to a digital analysis management system and a digital analysis management method based on the Internet.
Background
With the development of economy, shopping modes through E-commerce channels are more and more popular with people, but the sales volume of commodities are often not uniformly distributed and usually periodically changed, meanwhile, in the E-commerce shopping process, people pay more and more attention to objective comments of purchased users, and real user comments usually can make the users more desire to purchase.
In actual e-commerce transaction, in order to obtain the favorable comments of more users, a merchant often adopts some special activities to prompt the users to make the favorable comments as much as possible, but the comments generated under the condition are usually not objective comments, the comment contents are generally vague and derivative, and have no objectivity, the characteristics of the commodities cannot be clearly reflected in the comment contents, or the user directly copies the contents in the pre-comment to make the favorable comments, the mode often cannot effectively achieve the effect of attracting the user, and meanwhile, the comment also can interfere with the analysis of the commodity sales condition of the merchant, so that the merchant cannot accurately and effectively predict the commodity sales within a certain time later, and further the management of the merchant on the stock commodities is influenced.
Disclosure of Invention
The present invention is directed to a system and method for digital analysis and management based on the internet, so as to solve the problems in the background art.
In order to solve the technical problems, the invention provides the following technical scheme: an internet-based digital analysis management method, the method comprising the steps of:
s1, obtaining comment information respectively corresponding to commodities of different styles on an e-commerce platform and a commodity category to which each style commodity belongs;
s2, obtaining sales volumes corresponding to time lengths from the starting time of corresponding time periods to all time nodes in different time periods of each commodity type in historical data of the E-commerce platform, recording the sales volumes corresponding to the time lengths of the kth style commodity from the starting time of the nth time period to the ith time node in the nth time period as kni, analyzing the relation between the sales volumes of each commodity type and all the time nodes to obtain a first relation function, and taking data in the historical data every other first preset time length as a time period;
s3, obtaining each comment corresponding to each style commodity in the historical data within the latest first unit time, analyzing each comment corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
s4, according to the analysis results in the S2 and the S3, predicting the sales volume corresponding to each style commodity in the subsequent second unit time t2 based on the current time, and recording the time node corresponding to the current time in the time period as t1;
s5, the stock of each style commodity in the warehouse is obtained, the stock of each style commodity is compared with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time, and whether the stock of each style commodity is sufficient or not is judged.
Further, the method for analyzing the relationship between the sales volume of each commodity category and each time node in S2 includes the following steps:
s2.1, obtaining sales volumes corresponding to time nodes of each commodity type in different time periods in historical data of an e-commerce platform, and summarizing the sales volumes corresponding to the ith time node in the different time periods into a blank set which is marked as a sales volume data set corresponding to the ith time node;
s2.2, when i is different values, carrying out deviation screening on elements in the sales volume data set corresponding to the ith time node, recording the average value of the screened residual data as a reference sales volume XSi corresponding to the ith time node, and constructing a first relation data pair (i, XSi);
s2.3, constructing a plane rectangular coordinate system by taking o as an origin, a time node as an x axis and sales as a y axis, and marking coordinate points respectively corresponding to the (i, XSi) in the plane rectangular coordinate system by corresponding first relation data when the i is different values;
s2.4, extracting curve characteristic points of the marked coordinate points in the planar rectangular coordinate system, and connecting two adjacent curve characteristic points of the x-axis coordinate values in the extracted curve characteristic points to obtain a linear fitting result of the curve;
s2.5, comparing the relation between each curve characteristic point in the linear fitting result of the curve in the S2.4 and t1+ t2 respectively,
when the condition that the coordinate value of the x-axis in the characteristic point of the curve is more than or equal to t1 and less than or equal to t1+ t2 exists in the linear fitting result of the curve, judging that the extraction of the characteristic point of the curve is finished for the coordinate point marked in the plane rectangular coordinate system, wherein the function corresponding to the linear fitting result of the corresponding curve is a first relation function which is a piecewise function,
when the linear fitting result of the curve does not have the condition that the x-axis coordinate value in the curve characteristic point is more than or equal to t1 and less than or equal to t1+ t2, skipping to S2.4 to continue extracting the curve characteristic point on the basis of the currently obtained curve characteristic point;
the method for performing deviation screening on the sales volume data set corresponding to the ith time node in the S2.2 includes the following steps:
s2.2.1, obtaining a sales volume data set corresponding to the ith time node, recording a value corresponding to the mth element in the sales volume data set corresponding to the ith time node as Qim, recording an average value of values corresponding to all elements except the mth element in the sales volume data set corresponding to the ith time node as Qpim,
s2.2.2, determining the number of elements in the sales volume data set corresponding to the ith time node,
when the number of elements in the sales volume data set corresponding to the ith time node is equal to 1 or 2, judging that the elements in the sales volume data set corresponding to the ith time node are the result after deviation screening;
when the number of elements in the sales volume data set corresponding to the ith time node is greater than 2, calculating the deviation PLim between Qim and Qpim, wherein PLim = | Qim-Qpim |, and when m is obtained to be different values, recording the maximum value in each PLim as max { PLim },
if max { PLim } is larger than or equal to a first preset value in the database, removing elements corresponding to max { PLim } from the sales volume data set corresponding to the ith time node to obtain a new sales volume data set corresponding to the ith time node, skipping to S2.2.1 again to continuously obtain the new sales volume data set corresponding to the ith time node for deviation screening, wherein the first preset value is a preset constant in the database,
if max { PLim } is smaller than a first preset value in the database, taking an element in the sales volume data set corresponding to the ith time node as a result after deviation screening;
the method for extracting the curve characteristic points of the coordinate points marked in the rectangular plane coordinate system in the S2.4 comprises the following steps:
s2.4.1, obtaining the characteristic point of the curve obtained currently,
if the curve characteristic point obtained before is empty, selecting the marking point with the maximum x-axis coordinate value in the plane rectangular coordinate system as A, selecting the marking point with the minimum x-axis coordinate value in the plane rectangular coordinate system as B, taking A, B as two curve initial characteristic points,
if the currently acquired curve characteristic point is not empty, taking the currently acquired curve characteristic point as a curve initial characteristic point;
s2.4.2, connecting two curve initial characteristic points adjacent in the size of the x-axis coordinate value in the plane rectangular coordinate system to serve as a fitting line segment, and further obtaining a plurality of fitting line segments corresponding to the curve initial characteristic points, wherein the number of the fitting line segments is equal to the number of the curve initial characteristic points minus 1;
s2.4.3, selecting the points with the farthest distance from the m 1-th fitting line segment from the marking points in the plane rectangular system, and marking as Cm1, wherein the x-axis coordinate value in Cm1 is located between the x-axis coordinate values of the two curve characteristic points corresponding to the corresponding fitting line segments, and taking the Cm1 as a new curve characteristic point;
s2.4.4, summarizing curve initial characteristic points and new curve characteristic points extracted from the coordinate points marked in the planar rectangular coordinate system, and obtaining a result of extracting the curve characteristic points from the coordinate points marked in the planar rectangular coordinate system.
In the process of analyzing the relationship between the sales volume of each commodity category and each time node in the S2, the sales volume of the style commodity is considered to be changed periodically, and then data in different periods are respectively collected for analysis, so that the change condition of the sales volume of the style commodity along with the time node in one time period is obtained; when i is different values, performing deviation screening on elements in the sales volume data set corresponding to the ith time node, so as to eliminate elements with larger deviation caused by accidental factors in the sales volume data set corresponding to the ith time node, further enable the subsequently acquired first relation data pair to be more accurate, and provide data reference for acquiring the first relation function in the subsequent process; extracting curve characteristic points of coordinate points marked in the plane rectangular coordinate system, wherein the curve characteristic points are used for obtaining corresponding curve fitting results and providing data reference for obtaining a first relation function; comparing the relation between each curve characteristic point in the linear fitting result of the curve in the S2.4 and t1+ t2 respectively, so as to judge whether the accuracy of the linear fitting result of the corresponding curve is enough or not, and further judge whether to stop extracting the curve characteristic points from the coordinate points marked in the plane rectangular coordinate system or not; meanwhile, two parameters of t1 and t1+ t2 are selected as a standard for judging whether to stop curve characteristic point extraction on a coordinate point marked in the planar rectangular coordinate system, and the two parameters of t1 and t1+ t2 are also used as reference standards when the corresponding sales volume of each style commodity in the subsequent second unit time t2 based on the current time is predicted in the subsequent process.
Further, the method for acquiring the review comprehensive value of each style product in S3 includes the following steps:
s3.1, obtaining comments corresponding to each style commodity in the latest first unit time in the historical data and a commodity characteristic keyword library preset in a database;
s3.2, acquiring the number of good reviews and the number of bad reviews corresponding to each style commodity in the latest first unit time in the historical data, recording the number of good reviews corresponding to the kth style commodity in the latest first unit time in the historical data as HPk, recording the number of bad reviews corresponding to the kth style commodity in the latest first unit time in the historical data as CPk,
recording the j-th good comment of the kth style commodity in the latest first unit time as Pkj, wherein the interval duration of the time node corresponding to the comment with the large j value from the current time is small;
s3.3, extracting key words from Pkj, respectively matching the extracted key words with each type key word library preset in a database, wherein each type key word library comprises a tone word library, a commodity characteristic key word library, a tone word coefficient library and a grammar library, and obtaining a comment comprehensive value PZ corresponding to Pkj Pkj ；
S3.4, obtaining Pkj and its pre-evaluation, and marking as YL Pkj ；
S3.5, obtaining a comment comprehensive calibration value PZZ corresponding to Pkj Pkj ，
When YL Pkj If the value is larger than or equal to the first threshold value, judging that the jth comment of the kth style commodity in the latest first unit time isInvalid, pkj corresponding comment integrated calibration value PZZ Pkj =0，
When YL Pkj Less than a first threshold, PZZ Pkj =PZ Pkj *YL Pkj The first threshold is a preset constant in a database;
s3.6, obtaining a review comprehensive value PZK of the kth style commodity,
，
wherein E represents a review comprehensive value corresponding to one bad review preset in the database;
obtaining a comment comprehensive value PZ corresponding to Pkj Pkj The method comprises the following steps:
when the Pkj is subjected to keyword extraction, the comments are divided into different sentence nodes according to the sentence symbols as references, the keywords in the same sentence node are recorded into a blank set one by one according to the sequence of the keywords,
when the result of the Pkj keyword extraction is null, the comment comprehensive value PZ corresponding to Pkj is determined Pkj =0；
When the result of the Pkj keyword extraction is not empty, acquiring the corresponding linguistic word in the set corresponding to the e-th sentence node in Pkj, inquiring the corresponding basic coefficient of the corresponding linguistic word in the linguistic word coefficient library, marking as Re1Pkj,
acquiring the combination of the corresponding parts of speech of each keyword in the sequence from first to last in a set corresponding to the e sentence node in Pkj, comparing the combination result with a grammar library, screening out the grammar in the grammar library which is the same as the combination result, and obtaining the emotion adjusting coefficient, marked as Re2Pkj, corresponding to the screened grammar in a database,
acquiring the number of elements in the intersection of a set corresponding to the e-th sentence node in Pkj and the commodity characteristic keyword library, and recording the judgment result as Re3Pkj;
calculating Pkj corresponding comment comprehensive value
Wherein F represents the number of sentence nodes corresponding to Pkj,
max { Re3Pkj } represents the maximum value of Re3Pkj corresponding to the set corresponding to each sentence node in Pkj and the intersection of the commodity feature keyword libraries,
In the process of acquiring the comprehensive value of the comment of each style commodity in the S3, the comment is analyzed by combining the tone word library, the commodity characteristic keyword library, the tone word coefficient library and the grammar library in the database, and the comprehensive value of the comment corresponding to each comment is calculated; meanwhile, when considering that different user comments may be commented according to corresponding pre-comment references, and then the corresponding comments do not have client property, and further the comments do not have reference value, therefore, the corresponding comments need to be screened out, the screening process needs to consider the dependence value between the comments and the pre-comment references, and further data reference is provided for obtaining the comment comprehensive value of the kth style commodity.
Further, the method for acquiring the dependency value between Pkj and the preliminary review thereof in S3.4 comprises the following steps:
s3.4.1, obtaining all the previous comments in Pkj, said all previous comments being the previous j-1 favorable comments of the kth money item in the latest first unit time, the u th comment in the previous j-1 favorable comments of the kth money item in the latest first unit time is marked as the u th previous comment of Pkj,
when j =1, it is determined Pkj that there is no preceding comment, the dependency value YL between default Pkj and its preceding comment Pkj =1；
S3.4.2, acquiring the minimum information loading times W when the user browses the u-th front comment of Pkj from Pkj position uPkj The number of the comments which are loaded each time comment information is defaulted to be d, W in the database uPkj Is equal to the integer part of the quotient of u divided by d, u being differentW corresponding to the value uPkj The maximum value of (1) is denoted as W1 uPkj ；
S3.4.3, the same number of keywords between the u-th previous comment and Pkj for obtaining Pkj is noted as XT uPkj Respectively obtaining the minimum value of the total number of the u-th prepositive comment corresponding keywords of Pkj and the total number of the keywords corresponding to Pkj, and marking as GC uPkj The similarity XSD between the u-th preceding comment of Pkj and Pkj is obtained uPkj ，XSD uPkj =XT uPkj /GC uPkj ，GC uPkj ＞0；
S3.4.4, availability Pkj and its antecedent evaluation Pkj ，YL Pkj =max{XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]}，
The max { XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]Denotes u e [1,j-1 }]When u corresponds to XSD uPkj *（1-W uPkj /W1 uPkj ) Maximum value of (2).
In the process of obtaining the dependency value between Pkj and the preposed comment thereof, the load time pre-evaluation value between the preposed comment and Pkj and the similarity of the content between the preposed comment and Pkj are considered, and the greater the load time pre-evaluation value is, the lower the possibility that a user refers to the preposed comment is when the comment Pkj is made; the lower the similarity of the content between the pre-comment and Pkj, the lower the possibility that the user refers to the pre-comment when reviewing Pkj.
Further, the method for the sales corresponding to each style commodity in the subsequent second unit time t2 based on the current time in S4 includes the following steps:
s4.1, acquiring a first relation function and a comment comprehensive value PZK of the kth style commodity;
s4.2, substituting t1+ t2 into the independent variable in the first relation function, recording the corresponding function result as H (t 1+ t 2), substituting t1 into the independent variable in the first relation function, recording the corresponding function result as H (t 1), and obtaining the sales volumes H (t 1+ t 2) -H (t 1) of all the style commodities in the subsequent second unit time t2 based on the current time, wherein the commodity type corresponding to the kth style commodity corresponds to the commodity type;
s4.3, predicting the sales volume XSLk corresponding to the kth style commodity in the subsequent second unit time t2 based on the current time,
recording the minimum value of the comment comprehensive value of each style commodity in the commodity category corresponding to the kth style commodity as PZKX;
when PZKX is less than or equal to 0, judging
Wherein k1 represents the number of style commodities included in the commodity category corresponding to the kth style commodity, g is a balance coefficient, mk represents the cumulative sum of the review comprehensive values respectively corresponding to the style commodities included in the commodity category corresponding to the kth style commodity,
when PZKX > 0, then it is determined
In the process of the present invention based on the corresponding sales volume of each style commodity in the subsequent second unit time t2 of the current time, when PZKX is less than or equal to 0, g is set to balance PZKX and avoid the predicted XSLk being less than 0, and in this state, PZKX + g is greater than 0.
Further, in the process of obtaining the balance coefficient g, obtaining an independent variable substituting t1-t2 into the first relation function, recording a corresponding function result as H (t 1-t 2), obtaining an actual sales volume of the style goods corresponding to PZKX in the second unit time t2 before the current time as XSLSk, and obtaining the balance coefficient g, wherein g meets the condition:
further, the method for determining whether the stock quantity of each style commodity is sufficient in S5 includes the following steps:
s5.1, acquiring the inventory of each style commodity and the predicted value of the sales volume of the corresponding style commodity in the current time node;
s5.2, when the stock of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
An internet-based digital analysis management system, the system comprising the following modules:
the data acquisition module acquires comment information respectively corresponding to commodities of different styles on the E-commerce platform and the commodity category to which each style commodity belongs;
the sales data analysis module is used for acquiring sales corresponding to time lengths from the starting time of each commodity category to each time node in different time periods from the starting time of the corresponding time period to each time node in the historical data of the merchant platform, recording the sales corresponding to the time length from the starting time of the nth time period to the corresponding time length of the ith time node of the kth style commodity as kni, analyzing the relation between the sales of each commodity category and each time node to obtain a first relation function, and taking the data in the historical data every other first preset time length as a time period;
the comment analysis module is used for acquiring each comment corresponding to each style commodity in the latest first unit time in the historical data, analyzing each comment corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
the sales prediction module predicts the sales corresponding to each style commodity in the subsequent second unit time t2 based on the current time according to the analysis results in the sales data analysis module and the comment analysis module, and records the corresponding time node of the current time in the time period as t1;
and the inventory management module is used for acquiring the inventory of each style commodity in the warehouse, comparing the inventory of each style commodity with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time and judging whether the inventory of each style commodity is sufficient or not.
Further, in the process of judging whether the stock of each style commodity is sufficient or not, the stock management module obtains the stock of each style commodity and a predicted value of the sales volume of the corresponding style commodity in the current time node;
when the stock quantity of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, the stock quantity of the kth style commodity is judged to be sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
Compared with the prior art, the invention has the following beneficial effects: the invention not only can acquire the change condition of the sales volume of the commodity category along with the time node in the time period by analyzing the historical data, but also realizes the screening of invalid evaluation in the specific evaluation information of the commodity, avoids the interference of the invalid comments on the process of analyzing the sales condition of the commodity by a merchant, and accurately and effectively predicts the sales volume of the commodity in a certain time later by combining the quantized value of the rest evaluation after screening, thereby influencing the management of the merchant on the inventory commodity.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic diagram of the structure of an Internet-based digital analysis management system of the present invention;
fig. 2 is a flow chart of a digital analysis management method based on the internet.
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. 1-2, the present invention provides a technical solution: an internet-based digital analysis management method, the method comprising the steps of:
s1, obtaining comment information respectively corresponding to commodities of different styles on an e-commerce platform and a commodity category to which each style commodity belongs;
s2, obtaining sales volumes corresponding to time lengths from the starting time of corresponding time periods to all time nodes in different time periods of each commodity type in historical data of the E-commerce platform, recording the sales volumes corresponding to the time lengths of the kth style commodity from the starting time of the nth time period to the ith time node in the nth time period as kni, analyzing the relation between the sales volumes of each commodity type and all the time nodes to obtain a first relation function, and taking data in the historical data every other first preset time length as a time period;
in this embodiment, the first preset time duration is one year, and the time duration between two adjacent time nodes is recorded as a third preset value, which is one day;
the method for analyzing the relationship between the sales volume of each commodity category and each time node in the S2 comprises the following steps:
s2.1, obtaining sales volumes corresponding to time nodes of each commodity type in different time periods in historical data of the E-commerce platform, summarizing the sales volumes corresponding to the ith time node in different time periods into a blank set, and recording the blank set as a sales volume data set corresponding to the ith time node;
s2.2, when i is different values, performing deviation screening on elements in a sales volume data set corresponding to the ith time node, marking the average value of the screened residual data as a reference sales volume XSi corresponding to the ith time node, and constructing a first relation data pair (i, XSi);
s2.3, constructing a plane rectangular coordinate system by taking o as an origin, a time node as an x axis and sales as a y axis, and marking coordinate points respectively corresponding to the (i, XSi) in the plane rectangular coordinate system by corresponding first relation data when the i is different values;
s2.4, extracting curve characteristic points of the marked coordinate points in the planar rectangular coordinate system, and connecting two adjacent curve characteristic points of the x-axis coordinate values in the extracted curve characteristic points to obtain a linear fitting result of the curve;
s2.5, comparing the relation between each curve characteristic point in the linear fitting result of the curve in the S2.4 and t1+ t2 respectively,
when the condition that the coordinate value of the x-axis in the characteristic point of the curve is more than or equal to t1 and less than or equal to t1+ t2 exists in the linear fitting result of the curve, judging that the extraction of the characteristic point of the curve is finished for the coordinate point marked in the plane rectangular coordinate system, wherein the function corresponding to the linear fitting result of the corresponding curve is a first relation function which is a piecewise function,
when the linear fitting result of the curve does not have the condition that the x-axis coordinate value in the curve characteristic point is more than or equal to t1 and less than or equal to t1+ t2, skipping to S2.4 to continue extracting the curve characteristic point on the basis of the currently obtained curve characteristic point;
the method for performing deviation screening on the sales volume data set corresponding to the ith time node in the S2.2 comprises the following steps of:
s2.2.1, obtaining a sales volume data set corresponding to the ith time node, recording a value corresponding to the mth element in the sales volume data set corresponding to the ith time node as Qim, recording an average value of values corresponding to all elements except the mth element in the sales volume data set corresponding to the ith time node as Qpim,
s2.2.2, determining the number of elements in the sales volume data set corresponding to the ith time node,
when the number of elements in the sales volume data set corresponding to the ith time node is equal to 1 or 2, judging that the elements in the sales volume data set corresponding to the ith time node are the result after deviation screening;
when the number of elements in the sales volume data set corresponding to the ith time node is greater than 2, calculating the deviation PLim between Qim and Qpim, wherein PLim = | Qim-Qpim |, and when m is obtained to be different values, recording the maximum value in each PLim as max { PLim },
if max { PLim } is larger than or equal to a first preset value in the database, removing elements corresponding to max { PLim } from the sales volume data set corresponding to the ith time node to obtain a new sales volume data set corresponding to the ith time node, skipping to S2.2.1 again to continuously obtain the new sales volume data set corresponding to the ith time node for deviation screening, wherein the first preset value is a preset constant in the database,
if max { PLim } is smaller than a first preset value in the database, taking an element in the sales volume data set corresponding to the ith time node as a result after deviation screening;
in this embodiment, if the sales data set corresponding to the 1 st time node is {50, 200,230}, the first preset value is 120,
then Q11=50, qp11= (200 + 230) ÷ 2=215, pl11= |50-215| =165,
Q12=200，Qp12=（50+230）=140，PL12=|200-140|=60，
Q13=230，Qp13=（50+200）=125，PL13=|230-125|=105，
max { PLim } = {165,60,105} =165;
further, the element 50 corresponding to 165 is removed from the sales volume data set {50, 200,230} corresponding to the 1 st time node, and a new sales volume data set {200,230} corresponding to the first time node is obtained;
the method for extracting the curve characteristic points of the coordinate points marked in the plane rectangular coordinate system in the S2.4 comprises the following steps of:
s2.4.1, obtaining the characteristic point of the curve obtained at present,
if the curve characteristic point obtained before is empty, selecting the marking point with the maximum x-axis coordinate value in the plane rectangular coordinate system as A, selecting the marking point with the minimum x-axis coordinate value in the plane rectangular coordinate system as B, taking A, B as two curve initial characteristic points,
if the currently acquired curve characteristic point is not empty, taking the currently acquired curve characteristic point as a curve initial characteristic point;
s2.4.2, connecting two curve initial characteristic points adjacent in the size of the x-axis coordinate value in the plane rectangular coordinate system to serve as a fitting line segment, and further obtaining a plurality of fitting line segments corresponding to the curve initial characteristic points, wherein the number of the fitting line segments is equal to the number of the curve initial characteristic points minus 1;
s2.4.3, selecting the points with the farthest distance from the m 1-th fitting line segment from the marking points in the plane rectangular system, and marking as Cm1, wherein the x-axis coordinate value in Cm1 is located between the x-axis coordinate values of the two curve characteristic points corresponding to the corresponding fitting line segments, and taking the Cm1 as a new curve characteristic point;
s2.4.4, summarizing curve initial characteristic points and new curve characteristic points extracted from the marked coordinate points in the planar rectangular coordinate system to obtain a result of extracting the curve characteristic points from the marked coordinate points in the planar rectangular coordinate system.
S3, obtaining each comment corresponding to each style commodity in the historical data within the latest first unit time, analyzing each comment corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
the method for acquiring the comprehensive value of the comment of each style commodity in the S3 comprises the following steps:
s3.1, obtaining comments corresponding to each style commodity in the latest first unit time in the historical data and a commodity characteristic keyword library preset in a database;
s3.2, acquiring the number of good reviews and the number of bad reviews corresponding to each style commodity in the latest first unit time in the historical data, recording the number of good reviews corresponding to the kth style commodity in the latest first unit time in the historical data as HPk, recording the number of bad reviews corresponding to the kth style commodity in the latest first unit time in the historical data as CPk,
recording the jth good comment of the kth style commodity in the latest first unit time as Pkj, wherein the interval duration between the time node corresponding to the comment with the large j value and the current time is small;
s3.3, extracting key words from Pkj, respectively matching the extracted key words with each type key word library preset in a database, wherein each type key word library comprises a tone word library, a commodity characteristic key word library, a tone word coefficient library and a grammar library, and obtaining a comment comprehensive value PZ corresponding to Pkj Pkj ；
S3.4, obtaining the dependency value between Pkj and the preposed evaluation thereof, and marking the dependency value as YL Pkj ；
S3.5, obtaining a comment comprehensive calibration value PZZ corresponding to Pkj Pkj ，
When YL Pkj If the value is larger than or equal to the first threshold value, judging that the jth comment of the kth style commodity in the latest first unit time is invalid, and judging that the comment comprehensive calibration value PZZ corresponding to Pkj is invalid Pkj =0，
When YL Pkj Less than the first threshold, PZZ Pkj =PZ Pkj *YL Pkj The first threshold is a preset constant in a database;
s3.6, obtaining a review comprehensive value PZK of the kth style commodity,
，
wherein E represents a review comprehensive value corresponding to one bad review preset in the database;
obtaining a comment comprehensive value PZ corresponding to Pkj Pkj The method comprises the following steps:
when the Pkj is subjected to keyword extraction, the comments are divided into different sentence nodes according to the sentence symbols as references, the keywords in the same sentence node are recorded into a blank set one by one according to the sequence of the keywords,
when the result of the Pkj keyword extraction is null, the comment comprehensive value PZ corresponding to Pkj is determined Pkj =0；
When the result of Pkj extracting the keywords is not empty, acquiring the corresponding linguistic word in the set corresponding to the e-th sentence node in Pkj, inquiring the corresponding basic coefficient of the corresponding linguistic word in the linguistic word coefficient library, marking as Re1Pkj,
acquiring the combination of the corresponding parts of speech of each keyword in the sequence from first to last in a set corresponding to the e-th sentence node in Pkj, comparing the combination result with a grammar library, screening out the grammar in the grammar library which is the same as the combination result, obtaining the emotion adjusting coefficient corresponding to the screened grammar in the database, and marking the emotion adjusting coefficient as Re2Pkj,
acquiring the number of elements in the intersection of a set corresponding to the e-th statement node in Pkj and the commodity characteristic keyword library, and marking a judgment result as Re3Pkj;
calculating Pkj corresponding comment comprehensive value
Wherein F represents the number of sentence nodes corresponding to Pkj,
max { Re3Pkj } represents the maximum value of Re3Pkj corresponding to the set corresponding to each sentence node in Pkj and the intersection of the commodity feature keyword libraries,
The method for acquiring the dependency value between Pkj and the preliminary comment thereof in S3.4 comprises the following steps:
s3.4.1, obtaining all the previous comments in Pkj, said all previous comments being the previous j-1 favorable comments of the kth money item in the latest first unit time, the u th comment in the previous j-1 favorable comments of the kth money item in the latest first unit time is marked as the u th previous comment of Pkj,
when j =1, it is determined Pkj that there is no preceding comment, the dependency value YL between default Pkj and its preceding comment Pkj =1；
S3.4.2, acquiring the minimum information loading times W when the user browses the u-th front comment of Pkj from Pkj position uPkj The number of the comments loaded each time the comment information is loaded is d and W by default in the database uPkj Is equal to the integer part of the division of u by d, and when u is different values, the corresponding W is respectively uPkj The maximum value of (1) is denoted as W1 uPkj ；
S3.4.3, the same number of keywords between the u-th previous comment of Pkj and Pkj, which is recorded as XT uPkj Respectively obtaining the minimum value of the total number of the u th prepositive review corresponding keyword 5363 and the total number of the Pkj corresponding keyword of Pkj, and marking the minimum value as GC uPkj The similarity XSD between the u-th preceding comment of Pkj and Pkj is obtained uPkj ，XSD uPkj =XT uPkj /GC uPkj ，GC uPkj ＞0；
S3.4.4, availability Pkj and its antecedent evaluation Pkj ，YL Pkj =max{XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]}，
The max { XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]Denotes u e [1,j-1 }]When u corresponds to XSD uPkj *（1-W uPkj /W1 uPkj ) Maximum value of (2).
S4, according to the analysis results in the S2 and the S3, predicting the corresponding sales volume of each style commodity in the subsequent second unit time t2 based on the current time, and recording the corresponding time node of the current time in the time period as t1;
in S4, the method for obtaining the sales volume of each style commodity in the subsequent second unit time t2 based on the current time includes the following steps:
s4.1, acquiring a first relation function and a comment comprehensive value PZK of the kth style commodity;
s4.2, substituting t1+ t2 into the independent variable in the first relation function, marking the corresponding function result as H (t 1+ t 2), substituting t1 into the independent variable in the first relation function, marking the corresponding function result as H (t 1), and obtaining the sales volumes H (t 1+ t 2) -H (t 1) corresponding to all style commodities in the subsequent second unit time t2 based on the current time, of the commodity category corresponding to the kth style commodity;
s4.3, predicting the sales volume XSLk corresponding to the kth style commodity in the subsequent second unit time t2 based on the current time,
recording the minimum value of the comment comprehensive value of each style commodity in the commodity category corresponding to the kth style commodity as PZKX;
when PZKX is less than or equal to 0, judging
Wherein k1 represents the number of style commodities included in the commodity category corresponding to the kth style commodity, g is a balance coefficient, mk represents the cumulative sum of the review comprehensive values respectively corresponding to the style commodities included in the commodity category corresponding to the kth style commodity,
when PZKX > 0, then judge
In the process of obtaining the balance coefficient g, obtaining an independent variable substituting t1-t2 into the first relation function, recording a corresponding function result as H (t 1-t 2), obtaining the actual sales volume of the style commodity corresponding to PZKX in the second unit time t2 before the current time, and recording the actual sales volume as XSLSk, and obtaining the balance coefficient g, wherein g meets the condition:
s5, the stock of each style commodity in the warehouse is obtained, the stock of each style commodity is compared with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time, and whether the stock of each style commodity is sufficient or not is judged.
The method for judging whether the stock quantity of each style commodity is sufficient in the S5 comprises the following steps:
s5.1, acquiring the inventory of each style commodity and the predicted value of the sales volume of the corresponding style commodity in the current time node;
s5.2, when the stock of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
An internet-based digital analysis management system, the system comprising the following modules:
the data acquisition module acquires comment information respectively corresponding to commodities of different styles on the e-commerce platform and a commodity category to which each style commodity belongs;
the sales data analysis module is used for acquiring sales corresponding to time lengths from the starting time of each commodity category to each time node in different time periods from the starting time of the corresponding time period to each time node in the historical data of the merchant platform, recording the sales corresponding to the time length from the starting time of the nth time period to the corresponding time length of the ith time node of the kth style commodity as kni, analyzing the relation between the sales of each commodity category and each time node to obtain a first relation function, and taking the data in the historical data every other first preset time length as a time period;
the comment analysis module is used for acquiring each comment corresponding to each style commodity in the latest first unit time in the historical data, analyzing each comment corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
the sales prediction module predicts the sales corresponding to each style commodity in the subsequent second unit time t2 based on the current time according to the analysis results in the sales data analysis module and the comment analysis module, and records the corresponding time node of the current time in the time period as t1;
and the inventory management module is used for acquiring the inventory of each style commodity in the warehouse, comparing the inventory of each style commodity with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time, and judging whether the inventory of each style commodity is sufficient or not.
In the process of judging whether the stock of each style commodity is sufficient or not, the stock management module acquires the stock of each style commodity and a predicted value of the sales volume of the corresponding style commodity in the current time node;
when the stock quantity of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock quantity of the kth style commodity is sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. An internet-based digital analysis management method, comprising the steps of:
s1, obtaining comment information respectively corresponding to commodities of different styles on an e-commerce platform and a commodity category to which each style commodity belongs;
s2, obtaining sales volumes corresponding to time lengths from the starting time of corresponding time periods to all time nodes in different time periods of each commodity type in historical data of the E-commerce platform, recording the sales volumes corresponding to the time lengths of the kth style commodity from the starting time of the nth time period to the ith time node in the nth time period as kni, analyzing the relation between the sales volumes of each commodity type and all the time nodes to obtain a first relation function, and taking data in the historical data every other first preset time length as a time period;
s3, obtaining all comments corresponding to each style commodity in the latest first unit time in the historical data, analyzing all comments corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
s4, according to the analysis results in the S2 and the S3, predicting the corresponding sales volume of each style commodity in the subsequent second unit time t2 based on the current time, and recording the corresponding time node of the current time in the time period as t1;
s5, acquiring the stock of each style commodity in the warehouse, comparing the stock of each style commodity with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time, and judging whether the stock of each style commodity is sufficient;
the method for analyzing the relationship between the sales volume of each commodity category and each time node in the S2 comprises the following steps:
s2.1, obtaining sales volumes corresponding to time nodes of each commodity type in different time periods in historical data of the E-commerce platform, summarizing the sales volumes corresponding to the ith time node in different time periods into a blank set, and recording the blank set as a sales volume data set corresponding to the ith time node;
s2.2, when i is different values, carrying out deviation screening on elements in the sales volume data set corresponding to the ith time node, recording the average value of the screened residual data as a reference sales volume XSi corresponding to the ith time node, and constructing a first relation data pair (i, XSi);
s2.3, constructing a plane rectangular coordinate system by taking o as an origin, a time node as an x axis and sales as a y axis, and marking coordinate points respectively corresponding to the (i, XSi) in the plane rectangular coordinate system by corresponding first relation data when the i is different values;
s2.4, extracting curve characteristic points of the marked coordinate points in the planar rectangular coordinate system, and connecting two adjacent curve characteristic points of the x-axis coordinate values in the extracted curve characteristic points to obtain a linear fitting result of the curve;
s2.5, comparing the relation between each curve characteristic point in the linear fitting result of the curve in the S2.4 and t1+ t2 respectively,
when the condition that the coordinate value of the x-axis in the characteristic point of the curve is more than or equal to t1 and less than or equal to t1+ t2 exists in the linear fitting result of the curve, judging that the extraction of the characteristic point of the curve is finished for the coordinate point marked in the plane rectangular coordinate system, wherein the function corresponding to the linear fitting result of the corresponding curve is a first relation function which is a piecewise function,
when the linear fitting result of the curve does not have the condition that the x-axis coordinate value in the curve characteristic point is more than or equal to t1 and less than or equal to t1+ t2, skipping to S2.4 to continue extracting the curve characteristic point on the basis of the currently obtained curve characteristic point;
the method for performing deviation screening on the sales volume data set corresponding to the ith time node in the S2.2 includes the following steps:
s2.2.1, obtaining a sales volume data set corresponding to the ith time node, recording a value corresponding to the mth element in the sales volume data set corresponding to the ith time node as Qim, recording an average value of values corresponding to all elements except the mth element in the sales volume data set corresponding to the ith time node as Qpim,
s2.2.2, determining the number of elements in the sales volume data set corresponding to the ith time node,
when the number of elements in the sales volume data set corresponding to the ith time node is equal to 1 or 2, judging that the elements in the sales volume data set corresponding to the ith time node are the result of deviation screening;
when the number of elements in the sales volume data set corresponding to the ith time node is greater than 2, calculating the deviation PLim between Qim and Qpim, wherein PLim = | Qim-Qpim |, and when m is obtained to be different values, recording the maximum value in each PLim as max { PLim },
if max { PLim } is larger than or equal to a first preset value in the database, removing elements corresponding to max { PLim } from the sales volume data set corresponding to the ith time node to obtain a new sales volume data set corresponding to the ith time node, skipping to S2.2.1 again to continuously obtain the new sales volume data set corresponding to the ith time node for deviation screening, wherein the first preset value is a preset constant in the database,
if max { PLim } is smaller than a first preset value in the database, taking an element in the sales volume data set corresponding to the ith time node as a result after deviation screening;
the method for extracting the curve characteristic points of the coordinate points marked in the rectangular plane coordinate system in the S2.4 comprises the following steps:
s2.4.1, obtaining the characteristic point of the curve obtained currently,
if the currently acquired curve feature point is empty, selecting a mark point with the maximum x-axis coordinate value in the planar rectangular coordinate system as A, selecting a mark point with the minimum x-axis coordinate value in the planar rectangular coordinate system as B, taking A, B as two curve initial feature points,
if the currently acquired curve characteristic point is not empty, taking the currently acquired curve characteristic point as a curve initial characteristic point;
s2.4.2, connecting two curve initial characteristic points adjacent in the size of the x-axis coordinate value in the plane rectangular coordinate system to serve as a fitting line segment, and further obtaining a plurality of fitting line segments corresponding to the curve initial characteristic points, wherein the number of the fitting line segments is equal to the number of the curve initial characteristic points minus 1;
s2.4.3 selecting points which are respectively the farthest points from the m 1-th fitting line segment from the marking points in the plane rectangular system, and marking as Cm1, wherein the x-axis coordinate value in Cm1 is positioned between the x-axis coordinate values of the two curve characteristic points corresponding to the corresponding fitting line segments, and taking Cm1 as a new curve characteristic point;
s2.4.4, summarizing curve initial characteristic points and new curve characteristic points extracted from the marked coordinate points in the planar rectangular coordinate system to obtain a result of curve characteristic point extraction of the marked coordinate points in the planar rectangular coordinate system;
the method for acquiring the comprehensive comment value of each style commodity in the S3 comprises the following steps:
s3.1, obtaining comments corresponding to each style commodity in the latest first unit time in the historical data and a commodity characteristic keyword library preset in a database;
s3.2, acquiring the number of good reviews and the number of bad reviews corresponding to each style commodity in the latest first unit time in the historical data, recording the number of good reviews corresponding to the kth style commodity in the latest first unit time in the historical data as HPk, recording the number of bad reviews corresponding to the kth style commodity in the latest first unit time in the historical data as CPk,
recording the jth good comment of the kth style commodity in the latest first unit time as Pkj, wherein the interval duration between the time node corresponding to the comment with the large j value and the current time is small;
s3.3, extracting key words from Pkj, respectively matching the extracted key words with each type key word library preset in a database, wherein each type key word library comprises a tone word library, a commodity characteristic key word library, a tone word coefficient library and a grammar library, and obtaining a comment comprehensive value PZ corresponding to Pkj Pkj ；
S3.4, obtaining Pkj and its pre-evaluation, and marking as YL Pkj ；
S3.5, obtaining a comment comprehensive calibration value PZZ corresponding to Pkj Pkj ，
When YL Pkj If the value is larger than or equal to the first threshold value, judging that the jth comment of the kth style commodity in the latest first unit time is invalid, and judging that the comment comprehensive calibration value PZZ corresponding to Pkj is invalid Pkj =0，
When YL Pkj Less than a first threshold, PZZ Pkj =PZ Pkj *YL Pkj The first threshold is a preset constant in a database;
s3.6, obtaining a review comprehensive value PZK of the kth style commodity,
，
wherein E represents a review comprehensive value corresponding to one bad review preset in the database;
obtaining a comment comprehensive value PZ corresponding to Pkj Pkj The method comprises the following steps:
when the Pkj is subjected to keyword extraction, the comments are divided into different sentence nodes according to the sentence symbols as references, the keywords in the same sentence node are recorded into a blank set one by one according to the sequence of the keywords,
when the result of the Pkj keyword extraction is null, the comment comprehensive value PZ corresponding to Pkj is determined Pkj =0；
When the result of the Pkj keyword extraction is not empty, acquiring the corresponding linguistic word in the set corresponding to the e-th sentence node in Pkj, inquiring the corresponding basic coefficient of the corresponding linguistic word in the linguistic word coefficient library, marking as Re1Pkj,
acquiring the combination of the corresponding parts of speech of each keyword in the sequence from first to last in a set corresponding to the e-th sentence node in Pkj, comparing the combination result with a grammar library, screening out the grammar in the grammar library which is the same as the combination result, obtaining the emotion adjusting coefficient corresponding to the screened grammar in the database, and marking the emotion adjusting coefficient as Re2Pkj,
acquiring the number of elements in the intersection of a set corresponding to the e-th sentence node in Pkj and the commodity characteristic keyword library, and recording the judgment result as Re3Pkj;
calculating a comment comprehensive value Pkj
Wherein F represents the number of statement nodes corresponding to Pkj,
max { Re3Pkj } represents the maximum value of Re3Pkj corresponding to the set corresponding to each sentence node in Pkj and the intersection of the commodity feature keyword libraries,
the method for acquiring the dependency value between Pkj and the preliminary comment thereof in S3.4 comprises the following steps:
s3.4.1, obtaining all the previous comments in Pkj, said all previous comments being the previous j-1 favorable comments of the kth money item in the latest first unit time, the u th comment in the previous j-1 favorable comments of the kth money item in the latest first unit time is marked as the u th previous comment of Pkj,
when j =1, it is determined Pkj that there is no preceding comment, the dependency value YL between default Pkj and its preceding comment Pkj =1；
S3.4.2, get user to go from Pkj positionMinimum information load number W when visiting the u-th preceding comment of Pkj uPkj The number of the comments loaded each time the comment information is loaded is d and W by default in the database uPkj Is equal to the integer part of the division of u by d, and when u is different values, the corresponding W is respectively uPkj The maximum value of (1) is denoted as W1 uPkj ；
S3.4.3, the same number of keywords between the u-th previous comment of Pkj and Pkj, which is recorded as XT uPkj Respectively obtaining the minimum value of the total number of the u th prepositive review corresponding keyword 5363 and the total number of the Pkj corresponding keyword of Pkj, and marking the minimum value as GC uPkj The similarity XSD between the u-th preceding comment of Pkj and Pkj is obtained uPkj ，XSD uPkj =XT uPkj /GC uPkj ，GC uPkj ＞0；
S3.4.4, result Pkj dependency value YL between and its antecedent review Pkj ，YL Pkj =max{XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]}，
The max { XSD uPkj *（1-W uPkj /W1 uPkj ），u∈[1，j-1]Denotes u e [1,j-1 }]When u corresponds to XSD uPkj *（1-W uPkj /W1 uPkj ) Of (2) is calculated.
2. The internet-based digital analysis management method according to claim 1, wherein: in S4, the method for obtaining the sales volume of each style commodity in the subsequent second unit time t2 based on the current time includes the following steps:
s4.1, acquiring a first relation function and a comment comprehensive value PZK of the kth style commodity;
s4.2, substituting t1+ t2 into the independent variable in the first relation function, recording the corresponding function result as H (t 1+ t 2), substituting t1 into the independent variable in the first relation function, recording the corresponding function result as H (t 1), and obtaining the sales volumes H (t 1+ t 2) -H (t 1) of all the style commodities in the subsequent second unit time t2 based on the current time, wherein the commodity type corresponding to the kth style commodity corresponds to the commodity type;
s4.3, predicting the sales volume XSLk corresponding to the kth style commodity in the subsequent second unit time t2 based on the current time,
recording the minimum value of the comment comprehensive value of each style commodity in the commodity category corresponding to the kth style commodity as PZKX;
when PZKX is less than or equal to 0, judging
Wherein k1 represents the number of style commodities included in the commodity category corresponding to the kth style commodity, g is a balance coefficient, mk represents the cumulative sum of the review comprehensive values respectively corresponding to the style commodities included in the commodity category corresponding to the kth style commodity,
when PZKX > 0, then judge
3. The internet-based digital analysis management method according to claim 2, wherein: in the process of obtaining the balance coefficient g, obtaining an independent variable substituting t1-t2 into the first relation function, recording a corresponding function result as H (t 1-t 2), obtaining the actual sales volume of the style commodity corresponding to PZKX in the second unit time t2 before the current time, and recording the actual sales volume as XSLSk, and obtaining the balance coefficient g, wherein g meets the condition:
4. the internet-based digital analysis management method according to claim 1, wherein: the method for judging whether the stock quantity of each style commodity is sufficient in the S5 comprises the following steps:
s5.1, acquiring the inventory of each style commodity and the predicted value of the sales volume of the corresponding style commodity in the current time node;
s5.2, when the stock of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
5. An internet-based digital analysis management system to which the internet-based digital analysis management method according to any one of claims 1 to 4 is applied, the system comprising the following modules:
the data acquisition module acquires comment information respectively corresponding to commodities of different styles on the E-commerce platform and the commodity category to which each style commodity belongs;
the sales data analysis module is used for acquiring sales volumes corresponding to time durations from the starting time of corresponding time periods to all time nodes in different time periods for each commodity type in historical data of the merchant platform, recording the sales volume corresponding to the time duration from the starting time of the nth time period to the corresponding time duration of the ith time node of the kth style commodity as kni, analyzing the relationship between the sales volume of each commodity type and each time node to obtain a first relationship function, and taking data in every first preset time duration in the historical data as a time period;
the comment analysis module is used for acquiring each comment corresponding to each style commodity in the latest first unit time in the historical data, analyzing each comment corresponding to each style commodity by combining a commodity characteristic keyword library prefabricated in a database to obtain a comment comprehensive value of each style commodity, and recording the comment comprehensive value of the kth style commodity as PZK;
the sales prediction module predicts the sales corresponding to each style commodity in the subsequent second unit time t2 based on the current time according to the analysis results in the sales data analysis module and the comment analysis module, and records the corresponding time node of the current time in the time period as t1;
and the inventory management module is used for acquiring the inventory of each style commodity in the warehouse, comparing the inventory of each style commodity with the predicted value of the sales volume of the corresponding style commodity in the subsequent second unit time based on the current time, and judging whether the inventory of each style commodity is sufficient or not.
6. The internet-based digital analysis management system according to claim 5, wherein: in the process of judging whether the stock of each style commodity is sufficient or not, the stock management module acquires the stock of each style commodity and a predicted value of the sales volume of the corresponding style commodity in the current time node;
when the stock quantity of the kth style commodity is more than or equal to the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock quantity of the kth style commodity is sufficient,
and when the stock of the kth style commodity is less than the predicted value of the sales volume of the corresponding style commodity in the current time node, judging that the stock of the kth style commodity is insufficient, and early warning the administrator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211366741.2A CN115423575B (en) | 2022-11-03 | 2022-11-03 | Internet-based digital analysis management system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211366741.2A CN115423575B (en) | 2022-11-03 | 2022-11-03 | Internet-based digital analysis management system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115423575A CN115423575A (en) | 2022-12-02 |
| CN115423575B true CN115423575B (en) | 2023-02-14 |
Family
ID=84207542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211366741.2A Active CN115423575B (en) | 2022-11-03 | 2022-11-03 | Internet-based digital analysis management system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115423575B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116070995B (en) * | 2023-01-17 | 2023-10-24 | 深圳市秦丝科技有限公司 | Store purchase, sale and stock management system and method based on big data analysis |
| CN116362826A (en) * | 2023-03-01 | 2023-06-30 | 深圳市钱海网络技术有限公司 | Financial transaction information management system and method based on big data |
| CN116957709B (en) * | 2023-06-05 | 2024-04-23 | 上海朗晖慧科技术有限公司 | Manifest transaction management system and method based on digital cloud platform |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461876A (en) * | 2020-05-07 | 2020-07-28 | 赵玉洁 | E-commerce credit system management system and method based on big data |
| CN114022176A (en) * | 2021-09-26 | 2022-02-08 | 上海电信工程有限公司 | Method for predicting commodity sales on e-commerce platform and electronic equipment |
| CN114565344A (en) * | 2022-03-07 | 2022-05-31 | 泉州市米度信息科技有限公司 | ERP e-commerce order inventory management system and method based on cloud platform |
| CN114897575A (en) * | 2022-04-29 | 2022-08-12 | 广州简派供应链科技有限公司 | Intelligent sales platform based on big data and method thereof |
| CN115115315A (en) * | 2022-07-18 | 2022-09-27 | 深圳泽熙网络科技有限公司 | E-commerce commodity inventory quantity dynamic reminding management system based on cloud computing |
-
2022
- 2022-11-03 CN CN202211366741.2A patent/CN115423575B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115423575A (en) | 2022-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115423575B (en) | Internet-based digital analysis management system and method | |
| CN107016026B (en) | User tag determination method, information push method, user tag determination device, information push device | |
| TWI612488B (en) | Computer device and method for predicting market demand of commodities | |
| JP6177871B2 (en) | Disclosure of product information | |
| US10366435B2 (en) | Vehicle data system for rules based determination and real-time distribution of enhanced vehicle data in an online networked environment | |
| JP4920023B2 (en) | Inter-object competition index calculation method and system | |
| US20090198671A1 (en) | System and method for generating subphrase queries | |
| CN115115315A (en) | E-commerce commodity inventory quantity dynamic reminding management system based on cloud computing | |
| CN108153792A (en) | A kind of data processing method and relevant apparatus | |
| CN106649686B (en) | User interest grouping method and system based on the potential feature of multilayer | |
| CN111666757A (en) | Commodity comment emotional tendency analysis method, device and equipment and readable storage medium | |
| CN114997916A (en) | Prediction method, system, electronic device and storage medium of potential user | |
| KR20130038889A (en) | Object customization and management system | |
| Hu et al. | A meta-regression on the effect of online ratings on hotel room rates | |
| CN117668205A (en) | Smart logistics customer service processing method, system, equipment and storage medium | |
| CN114065063A (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| Huang et al. | Feature extraction of search product based on multi-feature fusion-oriented to Chinese online reviews | |
| CN117648491A (en) | Search system for electronic mall preferential merchants | |
| KR20150144916A (en) | system and method providing a suited shopping information by customer profiling | |
| CN112445905A (en) | Information processing method and device | |
| CN113781180B (en) | Article recommendation method and device, electronic equipment and storage medium | |
| JP2009053983A (en) | Information structurization apparatus, information structurization method and program | |
| US20110208738A1 (en) | Method for Determining an Enhanced Value to Keywords Having Sparse Data | |
| CN112328899A (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| CN113159927A (en) | Method and device for determining client label |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wang Hailong Inventor after: Chen Jian Inventor after: Liu Yong Inventor after: Liang Jian Inventor before: Chen Jian Inventor before: Liu Yong Inventor before: Liang Jian |
|
| CB03 | Change of inventor or designer information |