CN113112292A - Supervised commodity intelligent recommendation method and system in bulk commodity transaction - Google Patents
Supervised commodity intelligent recommendation method and system in bulk commodity transaction Download PDFInfo
- Publication number
- CN113112292A CN113112292A CN202110359552.1A CN202110359552A CN113112292A CN 113112292 A CN113112292 A CN 113112292A CN 202110359552 A CN202110359552 A CN 202110359552A CN 113112292 A CN113112292 A CN 113112292A
- Authority
- CN
- China
- Prior art keywords
- commodity
- transaction
- commodities
- data
- bulk
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title abstract description 23
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000010276 construction Methods 0.000 claims abstract description 4
- 238000012544 monitoring process Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 238000012952 Resampling Methods 0.000 claims description 2
- 230000002457 bidirectional effect Effects 0.000 claims description 2
- 238000013075 data extraction Methods 0.000 claims description 2
- 238000004140 cleaning Methods 0.000 claims 1
- 238000011524 similarity measure Methods 0.000 claims 1
- 230000006399 behavior Effects 0.000 abstract 1
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 14
- 230000000875 corresponding effect Effects 0.000 description 11
- 229910052742 iron Inorganic materials 0.000 description 7
- 229910000831 Steel Inorganic materials 0.000 description 5
- 239000010959 steel Substances 0.000 description 5
- 235000019482 Palm oil Nutrition 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 4
- 239000002540 palm oil Substances 0.000 description 4
- 238000011144 upstream manufacturing Methods 0.000 description 4
- CZMRCDWAGMRECN-UGDNZRGBSA-N Sucrose Chemical compound O[C@H]1[C@H](O)[C@@H](CO)O[C@@]1(CO)O[C@@H]1[C@H](O)[C@@H](O)[C@H](O)[C@@H](CO)O1 CZMRCDWAGMRECN-UGDNZRGBSA-N 0.000 description 3
- 229930006000 Sucrose Natural products 0.000 description 3
- 230000002596 correlated effect Effects 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 239000003549 soybean oil Substances 0.000 description 3
- 235000012424 soybean oil Nutrition 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013138 pruning Methods 0.000 description 2
- 240000000111 Saccharum officinarum Species 0.000 description 1
- 235000007201 Saccharum officinarum Nutrition 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000006467 substitution reaction Methods 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The invention discloses a supervised commodity intelligent recommendation method and system in bulk commodity transaction. The method is divided into four parts: the method comprises the steps of data processing of bulk commodity transaction data and market data, market data related commodity discovery, transaction data related commodity discovery, and construction of a related commodity set for intelligent recommendation and supervision of commodities. The invention solves the problems of fuzzy and uncertain commodity supervision range in the transaction supervision of bulk commodities. Firstly, data preprocessing operation is carried out on historical market information and transaction information of commodities. Secondly, the correlation of market data (price fluctuation and the like) of different commodities is analyzed to obtain price fluctuation related commodities. Then, the correlation of the transaction data (transaction behaviors and the like) of different commodities is analyzed to obtain the commodities related to the transaction events. And finally, combining the associated commodities formed in the two steps, constructing an associated commodity set, and intelligently recommending the supervised commodities according to the supervision task information.
Description
Technical Field
The invention belongs to the technical field of electronic commerce supervision of bulk commodities, and particularly relates to an intelligent supervision commodity recommendation method and system in bulk commodity transaction.
Background
In the course of electronic transaction supervision of bulk goods, the object of the supervision task usually includes the specific transaction goods. Many different products may be required in a supervision request initiated by a supervision party, and some other associated products may be involved in the supervision process. In a traditional supervision mode, a commodity supervision object is single and fixed, and the problems of incomplete supervision of commodities and the like are easy to occur; and the current electronic commerce market transaction mode of the bulk commodity is complex and changeable, and a plurality of risks exist in the transaction process, so that the electronic commerce market transaction mode is very necessary for monitoring the bulk commodity transaction. However, in the practical application process, the commodity object to be supervised is often set to be single and has an ambiguous direction, which aggravates the difficulty of checking the abnormal commodity, so that the related commodity of the abnormal commodity and the abnormality of the commodities on the upstream and downstream of the abnormal commodity are more difficult to be found. For example, in market trading, the increase in the price of iron ore may often be caused by price anomalies of its upstream and downstream commodities, which are difficult to detect in previous incomplete commodity regulations. Therefore, the intelligent recommendation method for supervising commodities is particularly important.
Disclosure of Invention
In order to solve the problems, the invention discloses a supervised commodity intelligent recommendation method and system in bulk commodity transaction, wherein the system is divided into four modules: the system comprises a data processing module for commodity transaction data and market data, a market data associated commodity discovery module, a transaction data associated commodity discovery module and an associated commodity set construction and supervision intelligent commodity recommendation module. Each module in the system is operated in a matched mode, a related commodity set is calculated according to supervision task information, historical market data and transaction data of supervised commodities, supervised commodities are recommended intelligently, and a supervision party can select proper bulk commodities to supervise the commodities better.
The main scheme is as follows:
the first part is data processing for bulk commodity transaction data and market data. In the actual transaction process of bulk commodities, the transaction price fluctuation of the commodities can cause a large amount of market data to be generated, including different price monthly lines, daily lines, minute lines and the like; and the data dimensions are various, including the highest price, the lowest price, the opening price, the closing price and the like in one day, so the actual price trend is very complex. In consideration of the actual characteristics of the price of the bulk commodity, firstly, corresponding data preprocessing operation is carried out according to the historical price time sequence of the bulk commodity. And then, in order to improve the readability of the market data and combine the actual situation of the large commodity trading market, selecting the daily closing price as the average price of the current day. And finally, extracting the traders and the trading commodities involved by the traders from the trading data, and constructing a bulk commodity trading transaction set.
The second part is to analyze the similarity of market data of different commodities and calculate the commodity related to price fluctuation. Firstly, according to the historical price time sequence of the commodities established in the last step, similarity measurement between different commodities is carried out. And selecting a Correlation Coefficient (namely Pearson Correlation Coefficient) as a mathematical tool, measuring the Correlation degree between the two variables, and reflecting whether the two variables are positively correlated or negatively correlated. The value range of the correlation coefficient is between [ -1,1], -1 represents absolute negative correlation, 1 represents absolute positive correlation, and 0 represents no correlation. And respectively measuring short-term price relevance r _ short and long-term price relevance r _ long between commodities, and simultaneously considering the two price relevance r _ long and r _ short to obtain a comprehensive correlation coefficient r _ sim of the commodity price fluctuation relation. And if the r _ sim meets the similarity threshold value tau, considering the pair of commodities as price fluctuation related commodities and adding the price fluctuation related commodities into the price fluctuation related commodity set.
And the third part is to analyze the correlation of the transaction data of different commodities and calculate the commodity related to the transaction event. Firstly, extracting corresponding transaction business and items corresponding to commodities from a large commodity transaction set constructed before. Subsequently, algorithm improvement is carried out on the basis of an association rule-based algorithm: and scanning the transaction set for the first time, calculating the support degree count of each commodity item, and pruning the commodity items which do not meet the support degree threshold value to construct a frequent item set. And then constructing a candidate two-item set according to the frequent one-item set, traversing the transaction set for the second time, pruning the items which do not meet the threshold value, and constructing the frequent two-item set. And finally, traversing the transaction set according to the commodities in the frequent binomial set, carrying out confidence calculation, defining the commodities meeting the support degree threshold and the confidence threshold simultaneously as transaction event associated commodities, and adding the commodities into the transaction event associated commodity set.
And the fourth part is to integrate the previously obtained price fluctuation related commodity set and the transaction event related commodity set and carry out intelligent recommendation of the commodity supervision object according to the supervision task requirement. And constructing a related commodity set by taking the price fluctuation related commodity set as a main part, and converting the same commodities appearing in the transaction event related commodity set into the related commodity set by considering the same commodities appearing in the transaction event related commodity set. Finally, according to the requirement of the initial supervision commodities in the supervision task, corresponding commodities are found in the associated commodity set and are recommended to the supervision party, and the purpose of intelligently recommending and supervising the commodities is achieved.
The invention analyzes the correlation of market data of different commodities to obtain the commodity associated with price fluctuation. Historical market information of the needed commodities is extracted from the database, corresponding data preprocessing operation is carried out on the information, similarity measurement is carried out on price fluctuation among different commodities, and the similarity meets a threshold value, namely the commodities are related to the price fluctuation. And secondly, analyzing the correlation of the transaction data of different commodities to obtain the commodities associated with the transaction event. And extracting the transaction information of the required commodity from the database, performing data processing to obtain a transaction set, and performing association rule algorithm analysis on the commodity to obtain the commodity which is associated with the transaction event if the association rule (namely the support degree and the confidence degree threshold value) is met. And finally, integrating the associated commodities formed in the two steps, constructing a bulk commodity association table, finding out the corresponding associated commodity in the commodity association table according to conditions such as user requirements and the like from the currently monitored commodity, and intelligently recommending the corresponding associated commodity to the supervision party. The method and the system mean that through the calculation of a software system, a supervision department can intelligently select commodity supervision objects according to supervision task information, historical transaction information, market information and the like when monitoring related commodities, so that the supervision quality and efficiency are greatly improved.
The invention has the beneficial effects that:
(1) the method and the system provided by the invention can be used for mining related market data and transaction data based on user requirements, effectively improving the rationality and the accuracy of commodity supervision range selection and increasing the success rate of finding abnormal commodities.
(2) Compared with the traditional commodity recommendation method based on association rules, the intelligent recommendation method provided by the invention can extract associated commodities only by constructing a frequent binomial set, simplifies the electronic transaction related data processing of bulk commodities, and ensures that the technical implementation process is clear and the operation is stable. And related commodity mining based on bulk commodity market data is added, the reliability of related commodity discovery is further ensured, market data related commodities can be obtained under the condition that some transaction data are lack, intelligent recommendation of monitored commodities is achieved, and the system has good dynamic adaptability. Meanwhile, each module between the systems is relatively independent, the data processing is simple and clear, and the fault tolerance is higher.
Drawings
FIG. 1 is a schematic diagram of the principal principle of the process of the present invention.
FIG. 2 is a schematic diagram of a discovery transaction event associated merchandise.
FIG. 3 is a schematic diagram of an article associated with a discovered price volatility.
Detailed Description
The present invention will be further illustrated with reference to the accompanying drawings and specific embodiments, which are to be understood as merely illustrative of the invention and not as limiting the scope of the invention. It should be noted that the terms "front," "back," "left," "right," "upper" and "lower" used in the following description refer to directions in the drawings, and the terms "inner" and "outer" refer to directions toward and away from, respectively, the geometric center of a particular component.
The intelligent supervision commodity recommendation system in the bulk commodity transaction of the embodiment comprises: (1) data processing
The first part of the method and the system is a data processing part, which extracts and processes market data and transaction data in the electronic transaction market of bulk commodities into usable data. As described above, the large commodity market transaction has the characteristics of large transaction quantity, large price fluctuation, rich commodity types, wide coverage scale and the like, and a large amount of market data including different commodity price monthly lines, daily lines, minute lines and the like are generated in the actual transaction process; and the data dimensions are various, including the highest price, the lowest price, the opening price, the closing price and the like in one day, so the actual price trend is very complex. Considering the relevant characteristics of the price of the bulk commodity and combining the actual situation of the market for trading the bulk commodity, it is necessary to perform corresponding data preprocessing operation on the historical price time sequence of the bulk commodity:
data extraction: and finding the historical price date K data of the commodity in the commodity market database.
Data filling: the vacancy values in the historical data are filled by the price of the previous day.
③ resampling: the closing price of the day is taken as the price of the day.
(2) Price volatility associated commodity discovery
And analyzing the similarity of price fluctuation among different commodities through the readable commodity price time sequence established in the last step to obtain the commodity associated with the price fluctuation. For example, analyzing price time series for item M and item N: xM＝[x1,x2,...,xt,...,xn]And XN＝[x1,x2,...,xt,...,xn]. Pearson Correlation Coefficient (Pearson Correlation Coefficient) was selected to measure the degree of Correlation between two time series of commodity prices. The specific correlation coefficient calculation method r (M, N) is shown by the following formula:
wherein XN(i) And XM(i) Representing the prices of the items N and M at the ith time point,
and
representing the average price of the items N and M. r (M, N) represents a correlation coefficient between the product M and the product N, i.e., similarity of price fluctuation, and is a correlated product when it exceeds a certain threshold. The similarity measurement of the method is divided into three steps:
analyzing a long-term correlation coefficient r _ long: and the correlation coefficient of the price sequences of the commodity M and the commodity N from the monitoring day to the present, namely the value of N in the formula is the maximum range recorded in the database.
Analyzing the short-term correlation coefficient r _ short: in the last 100 trading days, the correlation coefficient of the commodity M and the commodity N price sequence, namely the value of N in the above formula is 100.
And thirdly, simultaneously considering the r _ long and the r _ short to obtain a comprehensive correlation coefficient r _ sim of the commodity M and the commodity N.
The value range of the correlation coefficient is [ -1,1], r _ sim >0 represents positive correlation, r _ sim <0 represents negative correlation, and | r _ sim | represents the degree of correlation between variables. Usually, | r _ sim | is greater than some set threshold, the two variables are considered to have strong linear correlation. If the commodity M and the commodity N meet the corresponding correlation coefficient threshold values, the commodities are considered to be price fluctuation related commodities. Meanwhile, the specific correlation attribute can be judged according to the positive and negative of the correlation:
if the correlation coefficient r (M, N) of the commodity M and the commodity N exceeds a positive correlation threshold value, the two commodities are considered to show strong positive correlation, the price fluctuation shows the increase and decrease, and the commodities can be upstream and downstream commodities, complementary commodities or alternative commodities. Taking the actual production flow of steel as an example, the upstream and downstream commodities in the industrial chain are iron ore and deformed steel. The downstream of the iron ore is mainly derived from the requirement of the deformed steel. The price of iron ore rises, the production cost of the deformed steel bar increases, and the price of the deformed steel bar also rises. In another case, the two commodities have similar functions and have substitution effects, namely, the two commodities are substituted: for example, soybean oil and palm oil are a pair of substitutes, and as the price of soybean oil increases, the demand of soybean oil decreases and the demand of palm oil increases, so that the demand of palm oil increases and the price of palm oil also increases. And the price fluctuation relationship between the two commodities presents a strong positive correlation when the two commodities are combined together to be used, namely the commodities are complementary commodities.
And secondly, if the correlation coefficient r (M, N) of the commodity M and the commodity N exceeds a negative correlation threshold value, the two commodities are considered to show strong negative correlation, the price fluctuation shows that the correlation coefficient r (M, N) increases and decreases, and a complex industrial correlation relationship may exist between the commodities. For example, the correlation coefficient between white sugar and iron ore is r _ sim-0.3346, which becomes more obvious negative correlation. The possible reasons are: sugar cane as a raw material of white sugar is a warm-loving and bright-loving crop and is produced in northern hemisphere; iron ore is mainly produced in australia, major asia, etc. of the southern hemisphere; in the harvest season of the white sugar in northern hemisphere, the temperature in southern hemisphere is low, which is not beneficial to the exploitation of iron ore, so the yield is reduced. The difference in yield between the two results in negative similarity in price fluctuations. Of course, in addition to the reason and the present example, the commodities with strong negative correlation are often affected by complex factors of multiple factors such as climate, soil hydrology, international relations, sudden large-scale events and the like.
And analyzing the incidence relation between the appointed commodity and other commodities according to the algorithm, and constructing a price fluctuation incidence commodity set by taking the commodity with strong correlation as the incidence commodity of the commodity.
(3) Transaction event associated merchandise discovery
In the actual transaction process of a large number of commodities, besides market data, a large amount of transaction data is generated. The data relates to buyer information, seller information, transaction quantity, transaction time and other related information of the commodities, and the mining of the association relationship of the commodities is also a major key point for realizing intelligent recommendation and supervision of the commodity technology. (for example, the frequency of buying the combination of the A commodity and the C commodity by the trader is high, or the scale of the trading on the same day is similar, namely the related commodities.)
First, preprocessing of transaction data and information extraction are performed. Aiming at the technical scheme, effective transaction merchant data and commodity attribute data of buying and selling are selected, and the correlation of different commodity transaction data is analyzed. Secondly, on the basis of Apriori algorithm, a transaction event associated commodity discovery algorithm based on association rules is provided, and the algorithm is improved as follows: aiming at the characteristics of associated commodity supervision, extracting association rules when a frequent binomial set is constructed; ② the rule extraction is bidirectional. Thus, the following two concepts are redefined:
association rule (association rule): an association rule is an expression shaped as X → Y, where X and Y are different commodities. The strength of an association rule may be measured in terms of its support (support) and confidence (confidence). Support measures how often a rule can be used for a given data set, and confidence determines how often Y occurs in transactions that contain X.
Transaction event associated merchandise: if item X and item Y satisfy the corresponding support and confidence thresholds, an association rule in the form of X → Y or Y → X may be generated, i.e., they are considered a set of transaction event associated items.
Based on the above definitions and concepts, the solving algorithm for the trade event associated commodity discovery is shown as algorithm 1.
(4) Intelligent recommendation supervision commodity
And integrating the price fluctuation associated commodity set and the transaction event associated commodity set formed in the two steps to construct a total associated commodity table. In an actual market trading scenario, not all commodities are relevant to the trading event, but there is a correlation of price fluctuations among all commodities. Therefore, the construction of the associated commodity table is mainly based on the price fluctuation associated commodity set and assisted by the transaction event associated commodity set. If two incidence relations exist in a certain pair of commodities, the incidence degree of the commodities related to the transaction event is converted into the total incidence relation by taking the price fluctuation incidence degree as a standard.
And finally, according to the supervision task requirements, starting from the currently monitored related commodities, finding the corresponding commodities in the related commodity set, constructing a related commodity recommendation set, and recommending the related commodity recommendation set to a supervisor, so that the purpose of intelligently recommending and supervising the commodities is achieved.
The technical means disclosed in the invention scheme are not limited to the technical means disclosed in the above embodiments, but also include the technical scheme formed by any combination of the above technical features.
Claims (8)
1. A supervised commodity intelligent recommendation system in bulk commodity transaction is characterized by comprising four modules:
a first module: the data processing module is used for processing the transaction data and market data of bulk commodities;
a second module: the commodity discovery module is associated with the market quotation data of the bulk commodities;
a third module: a transaction data associated commodity discovery module;
a fourth module: an associated commodity set building and supervision commodity intelligent recommendation module;
each module in the system is operated in a matched mode, a related commodity set is calculated according to supervision task information, historical market data and transaction data of bulk commodities, and supervised commodities are recommended intelligently, so that the purpose of optimizing commodity supervision objects is achieved.
2. The intelligent supervised commodity recommendation system for bulk commodity transaction as recited in claim 1, wherein: and processing the transaction data of the bulk commodities: firstly, corresponding data preprocessing operation is carried out according to the historical price time sequence of the bulk commodity; then, in order to improve the readability of the market data and combine the actual situation of the large commodity trading market, selecting the closing price of each day as the average price of the day; and finally, extracting the traders and the trading commodities involved by the traders from the trading data, and constructing a bulk commodity trading transaction set.
3. The intelligent supervised commodity recommendation system for bulk commodity transaction as recited in claim 2, wherein: the bulk commodity transaction data processing module firstly performs data cleaning and only keeps data such as transaction merchant data and trade name of a transaction; selecting commodity data to establish a commodity set I so as to describe items and item sets of corresponding algorithms; item (1): one item for each commodity; item set: a collection of items, a set of items containing n items referred to as a set of n items; selecting traders therein to establish a trader set T so as to describe the affairs and the affair set of the corresponding algorithm; transaction: the trading behavior of each trader corresponds to one transaction; transaction set: a collection of all transactions.
4. The intelligent supervised commodity recommendation system for bulk commodity transaction as recited in claim 1, wherein said data processing module for market data comprises the following data preprocessing operations: data extraction: finding out historical price date K data of the commodities in a commodity market database; data filling: filling vacancy values in the historical data according to the price of the previous day; ③ resampling: taking the closing price closeprice of the current day as the price of the current day; finally, a readable historical price time series of the bulk good is established, for example, the price time series of the good N is: xN＝[x1,x2,...,xt,...,xn]。
5. The supervised commodity intelligent recommendation system in the bulk commodity transaction as recited in claim 4, wherein the bulk commodity market data associated commodity discovery module analyzes price fluctuation similarity among different commodities to obtain corresponding price fluctuation associated commodities according to a commodity historical price time sequence established by the market data processing module;
by analyzing the time series of prices of the commodities established in the previous step, such as the time series of prices of the commodity M and the commodity N: xM＝[x1,x2,...,xt,...,xn]And XN＝[x1,x2,...,xt,...,xn](ii) a Selecting Pearson Correlation CoefficPatent) that measures the degree of correlation between two time series of commodity prices. The specific correlation coefficient calculation method r (M, N) is shown by the following formula:
wherein XN(i) And XM(i) Representing the prices of the items N and M at the ith time point,
and
represents the average price of the items N and M; r (M, N) represents the correlation (similarity of price fluctuation) between the product M and the product N, and a product related to the similarity exceeding a certain threshold value is referred to as a related product.
6. The intelligent supervised commodity recommendation system for bulk commodity transaction as recited in claim 5, wherein the similarity measure is divided into three steps:
analyzing a long-term correlation coefficient r _ long: correlation coefficients between two commercial products since the date of monitoring (where n is the maximum range recorded in the database);
analyzing a short-term correlation coefficient r _ short: correlation coefficient between two commodities in 100 trading days (n is 100 in the above formula);
step three, simultaneously considering r _ long and r _ short to obtain a comprehensive correlation coefficient r _ sim of the commodity M and the commodity N;
the value range of the correlation coefficient is [ -1,1], r _ sim >0 represents positive correlation, r _ sim <0 represents negative correlation, and | r _ sim | represents the degree of correlation between variables. Usually, when | r _ sim | is larger than a certain set threshold, the two variables are considered to have strong linear correlation; if the commodity M and the commodity N meet the corresponding correlation coefficient threshold values, the commodities are considered to be price fluctuation related commodities.
7. The supervised commodity intelligent recommendation system in the bulk commodity transaction as recited in claim 3, wherein the bulk commodity transaction data associated commodity discovery module analyzes the transaction behavior correlation among different commodities according to the established commodity item set and the transaction behavior transaction set of traders to obtain transaction event associated commodities; on the basis of Apriori algorithm, a transaction event associated commodity discovery algorithm based on association rules is provided, and the algorithm is improved as follows: aiming at the characteristics of associated commodity supervision, extracting association rules when a frequent binomial set is constructed; the extraction of the rule is bidirectional; the definition of the transaction event associated merchandise is: if item X and item Y satisfy respective support and confidence thresholds, an association rule in the form of X → Y or Y → X may be generated, i.e., they are considered a set of transaction event associated items.
8. The intelligent supervised commodity recommendation system for bulk commodity transaction as recited in claim 1, wherein: the intelligent recommendation monitoring commodity: establishing a total associated commodity table by using the formed price fluctuation associated commodity set and the transaction event associated commodity set; the construction of the associated commodity table takes a price fluctuation associated commodity set as a main part and a transaction event associated commodity set as an auxiliary part; if two incidence relations exist in a certain pair of commodities, the incidence degree of the commodities related to the transaction event is converted into the total incidence relation by taking the price fluctuation incidence degree as a standard; and finally, according to the supervision task requirement, starting from the currently monitored related commodities, finding corresponding commodities in the related commodity set, constructing a related commodity recommendation set, and recommending the related commodity recommendation set to a supervisor, so that the purpose of intelligently recommending and supervising the commodities is achieved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110359552.1A CN113112292A (en) | 2021-04-02 | 2021-04-02 | Supervised commodity intelligent recommendation method and system in bulk commodity transaction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110359552.1A CN113112292A (en) | 2021-04-02 | 2021-04-02 | Supervised commodity intelligent recommendation method and system in bulk commodity transaction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113112292A true CN113112292A (en) | 2021-07-13 |
Family
ID=76713717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110359552.1A Pending CN113112292A (en) | 2021-04-02 | 2021-04-02 | Supervised commodity intelligent recommendation method and system in bulk commodity transaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113112292A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8374906B1 (en) * | 2008-09-30 | 2013-02-12 | Zilliant Incorporated | Method and system for generating pricing recommendations |
| CN107301592A (en) * | 2017-07-19 | 2017-10-27 | 北京京东尚科信息技术有限公司 | The method and device excavated for commodity substitute |
| CN107563857A (en) * | 2017-08-31 | 2018-01-09 | 齐鲁工业大学 | The customers buying behavior analysis method of logic-based reasoning negative customers rule trimming technology |
| CN108376347A (en) * | 2018-02-27 | 2018-08-07 | 广西财经学院 | A kind of commodity classification method based on A weighting priori algorithms |
| CN110910248A (en) * | 2019-11-22 | 2020-03-24 | 江苏云脑数据科技有限公司 | Bulk commodity futures arbitrage method, system and storage medium based on linear median regression |
-
2021
- 2021-04-02 CN CN202110359552.1A patent/CN113112292A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8374906B1 (en) * | 2008-09-30 | 2013-02-12 | Zilliant Incorporated | Method and system for generating pricing recommendations |
| CN107301592A (en) * | 2017-07-19 | 2017-10-27 | 北京京东尚科信息技术有限公司 | The method and device excavated for commodity substitute |
| CN107563857A (en) * | 2017-08-31 | 2018-01-09 | 齐鲁工业大学 | The customers buying behavior analysis method of logic-based reasoning negative customers rule trimming technology |
| CN108376347A (en) * | 2018-02-27 | 2018-08-07 | 广西财经学院 | A kind of commodity classification method based on A weighting priori algorithms |
| CN110910248A (en) * | 2019-11-22 | 2020-03-24 | 江苏云脑数据科技有限公司 | Bulk commodity futures arbitrage method, system and storage medium based on linear median regression |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021052031A1 (en) | Statistical interquartile range-based commodity inventory risk early warning method and system, and computer readable storage medium | |
| JP4358475B2 (en) | Credit evaluation system | |
| CN106021337A (en) | A big data analysis-based intelligent recommendation method and system | |
| CN102789618A (en) | Monitoring system and monitoring method for generating market business index | |
| CN110009502B (en) | Financial data analysis method, device, computer equipment and storage medium | |
| Cao et al. | A deep coupled LSTM approach for USD/CNY exchange rate forecasting | |
| CN109146611B (en) | E-commerce product quality evaluation method and system | |
| CN106780018A (en) | A kind of stock certificate data mining model reappeared based on history and system | |
| CN114022236A (en) | Multi-dimensional electric enterprise evaluation method | |
| CN109658154B (en) | Order prediction method based on regional trade data analysis | |
| CN116090915A (en) | Visual analysis method and system for enterprise market value analysis | |
| CN105303447A (en) | Method and device for carrying out credit rating through network information | |
| CN113112292A (en) | Supervised commodity intelligent recommendation method and system in bulk commodity transaction | |
| Yan et al. | Integrated data mining and topsis entropy weight method to evaluate logistics supply and demand efficiency of a 3PL company | |
| CN112711577A (en) | Abnormal transaction enterprise identification method and device | |
| Bhadury et al. | Constructing a Coincident Economic Indicator for India: How Well Does It Track Gross Domestic Product? | |
| Li et al. | A Medium to Long-Term Multi-Influencing Factor Copper Price Prediction Method Based on CNN-LSTM | |
| Shaoqian | A kind of improved algorithm for weighted Apriori and application to Data Mining | |
| CN113962486A (en) | Enterprise financial data prediction method and device, electronic equipment and storage medium | |
| KR102129491B1 (en) | Apparatus for analysing stocks | |
| CN114154617A (en) | Low-voltage resident user abnormal electricity utilization identification method and system based on VFL | |
| Chen et al. | An empirical study for the detection of corporate financial anomaly using outlier mining techniques | |
| Wu et al. | The volatility spillover effect between the ınternational crude oil futures price and China’s stock market-multivariate bekk-garch model based on wavelet multiresolution | |
| Ledion | Data mining techniques in database systems | |
| Kamley et al. | An Association Rule Mining Model for Finding the Interesting Patterns in Stock Market Dataset |
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 |