CN109583777A - A kind of financial product recommender system, method, equipment and medium - Google Patents
A kind of financial product recommender system, method, equipment and medium Download PDFInfo
- Publication number
- CN109583777A CN109583777A CN201811481623.XA CN201811481623A CN109583777A CN 109583777 A CN109583777 A CN 109583777A CN 201811481623 A CN201811481623 A CN 201811481623A CN 109583777 A CN109583777 A CN 109583777A
- Authority
- CN
- China
- Prior art keywords
- financial product
- cluster
- sample
- earning rate
- 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.)
- Pending
Links
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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23211—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with adaptive number of clusters
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Abstract
This application discloses a kind of financial product recommender system, method, equipment and media, comprising: quantitative value determining module, for determining present count magnitude N；Cluster centre determining module, for selecting N number of financial product sample from default financial product sample pool as N number of initial cluster center according to the present count magnitude N；Sample clustering module, for being based on N number of initial cluster center, and utilizes ISODATA algorithm, carries out cluster operation to all financial product samples in the default financial product sample pool, obtains corresponding cluster result；Sample determining module to be recommended, for determining multiple financial product samples to be recommended according to each final cluster centre in the cluster result；Communication module, for passing through preset communication interface for multiple financial product sample deliveries to be recommended to corresponding subscriber terminal equipment.The application can provide investment risk lower investment combination for user.
Description
Technical field
This application involves technical field of information recommendation, in particular to a kind of financial product recommender system, method, equipment and Jie
Matter.
Background technique
Currently, the assets for being held itself throwing is got used to reduce investment risk by individual investor or financial institution
It provides in multiple and different financial products, such as more stocks, multiple funds, more bonds, or investment stock market, fund and bond simultaneously
Etc. a variety of financial products.These investment combinations can reduce investment risk to a certain extent, and main thought is by making to select
The financial product selected is decentralized as far as possible, diversified to reach investment diversification person's investment risk, reduces unnecessary investment loss.
However, people are difficult to make preferably investment combination now.By taking the financial product stock of mainstream as an example, actual
During stock portfolio is chosen, since the quantity of personal share is big, type is huge more, and is easily influenced by stock market, therefore often
It is not easy to get more dispersed investment combination and selects stocks, people is caused to need to bear biggish investment risk.It how to be user
There is provided the lower investment combination of investment risk is to need further to be solved the problems, such as at present.
Summary of the invention
In view of this, the application's is designed to provide a kind of financial product recommender system, method, equipment and medium, energy
Enough provide investment risk lower investment combination for user.Its concrete scheme is as follows:
In a first aspect, this application discloses a kind of financial product recommender systems, comprising:
Quantitative value determining module, for determining present count magnitude N；
Cluster centre determining module, for being selected from default financial product sample pool according to the present count magnitude N
N number of financial product sample is as N number of initial cluster center；
Sample clustering module for being based on N number of initial cluster center, and utilizes ISODATA algorithm, to described default
All financial product samples in financial product sample pool carry out cluster operation, obtain corresponding cluster result；
Sample determining module to be recommended, for according to each final cluster centre in the cluster result, determine to
The multiple financial product samples recommended；
Communication module, for by preset communication interface by multiple financial product sample deliveries to be recommended to phase
The subscriber terminal equipment answered.
Optionally, the financial product recommender system, further includes:
Sample pool constructs module, for crawling the financial product in preset number of days from preset financial product database,
Obtain the default financial product sample pool.
Optionally, financial product recommender system, further includes:
Normalized module carries out the original daily earning rate of the financial product sample for utilizing preset formula
Normalized obtains normalizing daily earning rate accordingly；
Sample value determining module, for using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate the original daily earning rate, min table
Show that the original daily earning rate of minimum in the preset number of days, max indicate the original daily earning rate of maximum in the preset number of days.
Optionally, the quantitative value determining module, specifically for determining present count magnitude N using elbow method.
Second aspect, this application discloses a kind of financial product recommended methods, comprising:
Determine present count magnitude N；
According to the present count magnitude N, N number of financial product sample is selected from default financial product sample pool as N
A initial cluster center；
Based on N number of initial cluster center, and ISODATA algorithm is utilized, in the default financial product sample pool
All financial product samples carry out cluster operation, obtain corresponding cluster result；
According to each final cluster centre in the cluster result, multiple financial product samples to be recommended are determined；
By preset communication interface by multiple financial product sample deliveries to be recommended to corresponding user terminal
Equipment.
Optionally, before the determining present count magnitude N, further includes:
The financial product in preset number of days is crawled from preset financial product database, obtains the default financial product
Sample pool.
Optionally, before the determining present count magnitude N, further includes:
Using preset formula, the original daily earning rate of the financial product sample is normalized, is obtained corresponding
Normalization daily earning rate；
Using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate the original daily earning rate, min table
Show that the original daily earning rate of minimum in the preset number of days, max indicate the original daily earning rate of maximum in the preset number of days.
Optionally, the determining present count magnitude N, comprising:
Present count magnitude N is determined using elbow method.
The third aspect, this application discloses a kind of financial product recommendation apparatus, comprising:
Memory, for saving computer program；
Processor, for executing the computer program, to realize aforementioned disclosed financial product recommended method.
Fourth aspect, this application discloses a kind of computer readable storage mediums, for saving computer program, wherein
The computer program realizes aforementioned disclosed financial product recommended method when being executed by processor.
As it can be seen that the application after determining present count magnitude, selects respective counts from default financial product sample pool
Then the financial product sample of amount gathers all financial product samples using ISODATA algorithm as initial cluster center
Generic operation determines financial product sample to be recommended, and lead to then according to each final cluster centre in cluster result
It crosses default communication interface and recommends to corresponding subscriber terminal equipment.It can be seen that the application is based on ISODATA algorithm to default
All financial product samples in financial product sample pool carry out cluster operation, in multiple final clusters available in this way
The heart, and the similitude between these cluster centres is very low, with the lower multiple cluster centres of these similitudes for according to institute
The multiple financial product samples to be recommended determined have bigger dispersibility, advantageously reduce the investment wind of user in this way
Danger.To sum up, the application can provide investment risk lower investment combination for user.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of application for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of financial product recommender system structural schematic diagram disclosed in the present application；
Fig. 2 is elbow method application schematic diagram；
Fig. 3 is a kind of financial product recommended method flow chart disclosed in the present application；
Fig. 4 is a kind of specific financial product recommended method flow chart disclosed in the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
The embodiment of the present application discloses a kind of financial product recommender system, shown in Figure 1, which includes:
Quantitative value determining module 11, for determining present count magnitude N；
Cluster centre determining module 12, for being chosen from default financial product sample pool according to the present count magnitude N
N number of financial product sample is as N number of initial cluster center out；
Sample clustering module 13, for being based on N number of initial cluster center, and (i.e. using ISODATA algorithm
Iterative Selforganizing Data Analysis Techniques Algorithm, iteration self-organizing data point
Analyse algorithm), cluster operation is carried out to all financial product samples in the default financial product sample pool, is gathered accordingly
Class result；
Sample determining module 14 to be recommended, for determining according to each final cluster centre in the cluster result
Multiple financial product samples to be recommended；
Communication module 15, for by preset communication interface by multiple financial product sample deliveries to be recommended extremely
Corresponding subscriber terminal equipment.
It is understood that can specifically pass through Europe during carrying out above-mentioned cluster operation using ISODATA algorithm
Formula distance calculates the similarity between each financial product and each cluster centre, in addition, during above-mentioned cluster operation, also
Cluster centre can be corrected and calculate the range index function of each financial product sample in all kinds of, and according to given requirement, it will
The preceding cluster set once obtained merges and division processing, so that new cluster centre is obtained, particularly, if cluster centre
Number is less than or equal to the half of desired value, and the number of interative computation is that the number of odd-times iteration and cluster centre is small in other words
In twice of desired value, then division processing is carried out to existing cluster；On the contrary, if the number of interative computation is even-times or gathers
The number at class center is greater than twice of desired value, then carries out corresponding merging treatment.In addition, need to also fall into a trap in each iterative process
Parameters index is calculated, and judges whether cluster result meets the requirements according to obtained parameter index.After successive ignition,
If result restrains, clustering operation terminates, to obtain corresponding cluster result.
Further, the sample determining module 14 to be recommended is when determining financial product sample to be recommended, if cluster
As a result the cluster centre of middle clustering cluster is some financial product sample, then directly choose the cluster centre as investment combination to
The financial product sample of recommendation；If the cluster centre of clustering cluster is not financial product sample data, Euclidean distance method meter is used
Calculate the distance of all financial product sample datas in the cluster to cluster centre, and the smallest financial product sample conduct of selected distance
Financial product sample to be recommended.
It is understood that the financial product in the present embodiment includes but is not limited to stock, fund, bond, futures, declaration form
Deng.
As it can be seen that the embodiment of the present application after determining present count magnitude, is selected from default financial product sample pool
The financial product sample of respective numbers is as initial cluster center, then using ISODATA algorithm to all financial product samples
It carries out cluster operation and determines financial product sample to be recommended then according to each final cluster centre in cluster result
This, and corresponding subscriber terminal equipment is recommended to by default communication interface.It can be seen that the embodiment of the present application is to be based on
ISODATA algorithm carries out cluster operation to all financial product samples in default financial product sample pool, available in this way
Multiple final cluster centres, and the similitude between these cluster centres is very low, it is lower multiple with these similitudes
Cluster centre is conducive to drop in this way to have bigger dispersibility according to the multiple financial product samples to be recommended determined
The investment risk of low user.To sum up, the embodiment of the present application can provide investment risk lower investment combination for user.
On the basis of previous embodiment, the embodiment of the present application has made further optimization and explanation to technical solution.Tool
Body:
In the present embodiment, the financial product recommender system be can further include:
Sample pool constructs module, for crawling the financial product in preset number of days from preset financial product database,
Obtain the default financial product sample pool.
For example, default day can be crawled from the common stock pond such as 300 constituent stocks of Shanghai and Shenzhen, Index of Shanghai Stock Exchange, Shenzhen Index
Then the data crawled are added in default financial product sample pool by the stock exchange data in number.In the present embodiment, on
The time for the process of crawling is stated as unit of number of days, wherein the selection of number of days has strong influence, number of days mistake to the effect of cluster
The long computation complexity that will increase cluster process, the too short accuracy and confidence level that can reduce cluster of number of days, by opinion repeatedly
Card, the present embodiment can be range locating for above-mentioned preset number of days 30 to 60 days.
Further, due to consideration that the financial products such as different stocks have different dimensions, this can be to data similarity
Result have an impact, in order to eliminate the dimension impact between daily earning rate, calculate the similarity between each financial product it
Before, need first to be standardized the financial products such as stock, fund, with solve between each financial product daily earning rate can
Than property problem.Specifically, the financial product recommender system, can also include:
Normalized module carries out the original daily earning rate of the financial product sample for utilizing preset formula
Normalized obtains normalizing daily earning rate accordingly；
Sample value determining module, for using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate the original daily earning rate, min table
Show that the original daily earning rate of minimum in the preset number of days, max indicate the original daily earning rate of maximum in the preset number of days.
Wherein, the calculation formula of original daily earning rate P_value specifically:
P_value=(incomei-incomei-1)/incomei-1；
Wherein, incomeiRefer to i-th day income of financial product.Then original daily earning rate P_value indicates same day income
Compared to the growth rate of income yesterday.Therefore each financial product can be regarded as within the set time by gold on the basis of earning rate
Melt time series.Here it chooses earning rate and carries out analysis as measurement and be derived from earning rate and be more able to reflect the financial products such as stock
Variation tendency, it is with a high credibility, and calculate simple, be easily obtained.
It is understood that raw financial product data are by after above-mentioned normalized, the day of each financial product is received
Beneficial rate is in the same order of magnitude so that it is convenient in the calculating for the difference for carrying out the daily earning rate between different financial products.This implementation
In example, the daily earning rate deviation of two financial products is the important finger for judging the correlation of the two financial products in specific time
Mark.
In addition, in the present embodiment, the quantitative value determining module specifically can be used for determining preset quantity using elbow method
Value N.
Wherein, elbow method is used to determine that the core concept of the number of initial cluster center to be: with the increasing of cluster numbers k
Greatly, financial product sample can be divided more fine, and the cluster degree of each cluster can be gradually increased, such error sum of squares
SSE will gradually become smaller naturally.Also, when cluster numbers k be less than true cluster numbers when, due to the increase of k will increase dramatically it is each
The extent of polymerization of cluster, therefore the fall of SSE can be very big, and when k reaches true cluster numbers, it is further added by the obtained polymerization of k
Degree return can become smaller rapidly, so fall can die-off, then as k value continuing increase and tend towards stability, also
It is to say that the relational graph of SSE and k is the shape of an elbow, it is shown in Figure 2, and the corresponding k value of this ancon is exactly data
True cluster numbers.By taking example shown in Figure 2 as an example, ancon for k value be 4, therefore for cluster data corresponding to Fig. 2
For, optimal initial clustering number should be 4.
It should be pointed out that in order to calculate above-mentioned error sum of squares SSE, it can be random from default financial product sample pool
K cluster centre is selected, fixed value such as 10 from 2 to one k is enabled by enumerating, then reruns for several times in each k value
K-means algorithm, and the corresponding SSE of current k is calculated, specific calculation formula is as follows:
Wherein, CiIndicate that i-th of cluster, p are cluster CiSample point, miIt is cluster CiIn all samples mean value, SSE is all
The cluster error of sample and the core index of elbow method, represent the quality of Clustering Effect.
As it can be seen that the quantity of the initial cluster center in the present embodiment in ISODATA cluster process is by multiple k-means
And in conjunction with elbow method come what is determined, the number of iterations of clustering algorithm is advantageously reduced in this way, the time required to reducing calculating process.Separately
Outside, the present embodiment is effectively avoided by the normalized of data causes data dimension variant in turn because capital is different
The case where influencing cluster result, occurs.
Shown in Figure 3, the embodiment of the present application further discloses a kind of financial product recommended method, comprising:
Step S11: present count magnitude N is determined；
Step S12: according to the present count magnitude N, N number of financial product sample is selected from default financial product sample pool
This is as N number of initial cluster center；
Step S13: it is based on N number of initial cluster center, and utilizes ISODATA algorithm, to the default financial product
All financial product samples in sample pool carry out cluster operation, obtain corresponding cluster result；
Step S14: according to each final cluster centre in the cluster result, determine that multiple finance to be recommended produce
Product sample；
Step S15: by preset communication interface by multiple financial product sample deliveries to be recommended to corresponding
Subscriber terminal equipment.
It is understood that can specifically pass through Europe during carrying out above-mentioned cluster operation using ISODATA algorithm
Formula distance calculates the similarity between each financial product and each cluster centre, in addition, during above-mentioned cluster operation, also
Cluster centre can be corrected and calculate the range index function of each financial product sample in all kinds of, and according to given requirement, it will
The preceding cluster set once obtained merges and division processing, so that new cluster centre is obtained, particularly, if cluster centre
Number is less than or equal to the half of desired value, and the number of interative computation is that the number of even-times iteration and cluster centre is small in other words
In twice of desired value, then division processing is carried out to existing cluster；On the contrary, if the number of interative computation is odd-times or gathers
The number at class center is greater than twice of desired value, then carries out corresponding merging treatment.In addition, need to also fall into a trap in each iterative process
Parameters index is calculated, and judges whether cluster result meets the requirements according to obtained parameter index.After successive ignition,
If result restrains, clustering operation terminates, to obtain corresponding cluster result.
Further, when determining financial product sample to be recommended, if the cluster centre of clustering cluster is in cluster result
Some financial product sample then directly chooses to be recommended financial product sample of the cluster centre as investment combination；If poly-
The cluster centre of class cluster is not financial product sample data, then calculates all financial product samples in the cluster with Euclidean distance method
Data are to the distance of cluster centre, and the smallest financial product sample of selected distance is as financial product sample to be recommended.
It is understood that the financial product in the present embodiment includes but is not limited to stock, fund, bond, futures, declaration form
Deng.
As it can be seen that the embodiment of the present application after determining present count magnitude, is selected from default financial product sample pool
The financial product sample of respective numbers is as initial cluster center, then using ISODATA algorithm to all financial product samples
It carries out cluster operation and determines financial product sample to be recommended then according to each final cluster centre in cluster result
This, and corresponding subscriber terminal equipment is recommended to by default communication interface.It can be seen that the embodiment of the present application is to be based on
ISODATA algorithm carries out cluster operation to all financial product samples in default financial product sample pool, available in this way
Multiple final cluster centres, and the similitude between these cluster centres is very low, it is lower multiple with these similitudes
Cluster centre is conducive to drop in this way to have bigger dispersibility according to the multiple financial product samples to be recommended determined
The investment risk of low user.To sum up, the embodiment of the present application can provide investment risk lower investment combination for user.
Further, before the determining present count magnitude N, can also include:
The financial product in preset number of days is crawled from preset financial product database, obtains the default financial product
Sample pool.
For example, default day can be crawled from the common stock pond such as 300 constituent stocks of Shanghai and Shenzhen, Index of Shanghai Stock Exchange, Shenzhen Index
Then the data crawled are added in default financial product sample pool by the stock exchange data in number.In the present embodiment, on
The time for the process of crawling is stated as unit of number of days, wherein the selection of number of days has strong influence, number of days mistake to the effect of cluster
The long computation complexity that will increase cluster process, the too short accuracy and confidence level that can reduce cluster of number of days, by opinion repeatedly
Card, the present embodiment can be range locating for above-mentioned preset number of days 30 to 60 days.
Further, due to consideration that the financial products such as different stocks have different dimensions, this can be to data similarity
Result have an impact, in order to eliminate the dimension impact between daily earning rate, calculate the similarity between each financial product it
Before, need first to be standardized the financial products such as stock, fund, with solve between each financial product daily earning rate can
Than property problem.Specifically, can also include: before the determining present count magnitude N
Using preset formula, the original daily earning rate of the financial product sample is normalized, is obtained corresponding
Normalization daily earning rate；
Using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate the original daily earning rate, min table
Show that the original daily earning rate of minimum in the preset number of days, max indicate the original daily earning rate of maximum in the preset number of days.
Wherein, the calculation formula of original daily earning rate P_value specifically:
P_value=(incomei-incomei-1)/incomei-1；
Wherein, incomeiRefer to i-th day income of financial product.Then original daily earning rate P_value indicates same day income
Compared to the growth rate of income yesterday.Therefore each financial product can be regarded as within the set time by gold on the basis of earning rate
Melt time series.Here it chooses earning rate and carries out analysis as measurement and be derived from earning rate and be more able to reflect the financial products such as stock
Variation tendency, it is with a high credibility, and calculate simple, be easily obtained.
It is understood that raw financial product data are by after above-mentioned normalized, the day of each financial product is received
Beneficial rate is in the same order of magnitude so that it is convenient in the calculating for the difference for carrying out the daily earning rate between different financial products.This implementation
In example, the daily earning rate deviation of two financial products is the important finger for judging the correlation of the two financial products in specific time
Mark.
Further, the determining present count magnitude N, can specifically include: determine present count magnitude N using elbow method.
Wherein, elbow method is used to determine that the core concept of the number of initial cluster center to be: with the increasing of cluster numbers k
Greatly, financial product sample can be divided more fine, and the cluster degree of each cluster can be gradually increased, such error sum of squares
SSE will gradually become smaller naturally.Also, when cluster numbers k be less than true cluster numbers when, due to the increase of k will increase dramatically it is each
The extent of polymerization of cluster, therefore the fall of SSE can be very big, and when k reaches true cluster numbers, it is further added by the obtained polymerization of k
Degree return can become smaller rapidly, so fall can die-off, then as k value continuing increase and tend towards stability, also
It is to say that the relational graph of SSE and k is the shape of an elbow, and the corresponding k value of this ancon is exactly the true cluster numbers of data.
It should be pointed out that in order to calculate above-mentioned error sum of squares SSE, it can be random from default financial product sample pool
K cluster centre is selected, fixed value such as 10 from 2 to one k is enabled by enumerating, then reruns for several times in each k value
K-means algorithm, and the corresponding SSE of current k is calculated, specific calculation formula is as follows:
Wherein, CiIndicate that i-th of cluster, p are cluster CiSample point, miIt is cluster CiIn all samples mean value, SSE is all
The cluster error of sample and the core index of elbow method, represent the quality of Clustering Effect.
As it can be seen that the quantity of the initial cluster center in the present embodiment in ISODATA cluster process is by multiple k-means
And in conjunction with elbow method come what is determined, the number of iterations of clustering algorithm is advantageously reduced in this way, the time required to reducing calculating process.Separately
Outside, the present embodiment is effectively avoided by the normalized of data causes data dimension variant in turn because capital is different
The case where influencing cluster result, occurs.
Shown in Figure 4, the embodiment of the present application discloses a kind of specific stock recommended method, comprising:
Step S21: crawling the stock sample in preset number of days from preset stock database, obtains default stock sample
Pond.
Step S22: the normalization daily earning rate of each stock in the default stock sample pool is calculated, and as stock
The sample value of tickets sample sheet.
Step S23: present count magnitude N is determined using elbow method.
Step S24: according to the present count magnitude N, N number of stock sample is randomly selected out from default stock sample pool and is made
For N number of initial cluster center.
Step S25: it is based on N number of initial cluster center, and utilizes ISODATA algorithm, to the default stock sample
All stock samples in pond carry out cluster operation, obtain corresponding cluster result.
Step S26: according to each final cluster centre in the cluster result, multiple stock samples to be recommended are determined
This.
Step S27: by preset communication interface by multiple stock sample deliveries to be recommended to corresponding user
Terminal device.
Further, the embodiment of the present application also discloses a kind of financial product recommendation apparatus, comprising:
Memory, for saving computer program；
Processor, for executing the computer program, to realize financial product recommended method disclosed in previous embodiment.
Further, the embodiment of the present application also discloses a kind of computer readable storage medium, for saving computer journey
Sequence, wherein previous embodiment discloses financial product recommended method when the computer program is executed by processor.
It wherein, can be with reference to corresponding contents disclosed in previous embodiment, herein not about the specific steps of the above method
It is repeated again.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Detailed Jie has been carried out to a kind of financial product recommender system, method, equipment and medium provided herein above
It continues, specific examples are used herein to illustrate the principle and implementation manner of the present application, and the explanation of above embodiments is only
It is to be used to help understand the method for this application and its core ideas；At the same time, for those skilled in the art, according to this Shen
Thought please, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not manage
Solution is the limitation to the application.
Claims (10)
1. a kind of financial product recommender system characterized by comprising
Quantitative value determining module, for determining present count magnitude N；
Cluster centre determining module, for being selected from default financial product sample pool N number of according to the present count magnitude N
Financial product sample is as N number of initial cluster center；
Sample clustering module for being based on N number of initial cluster center, and utilizes ISODATA algorithm, to the default finance
All financial product samples in product sample pond carry out cluster operation, obtain corresponding cluster result；
Sample determining module to be recommended, for determining to be recommended according to each final cluster centre in the cluster result
Multiple financial product samples；
Communication module, for by preset communication interface by multiple financial product sample deliveries to be recommended to corresponding
Subscriber terminal equipment.
2. financial product recommender system according to claim 1, which is characterized in that further include:
Sample pool building module is obtained for crawling the financial product in preset number of days from preset financial product database
The default financial product sample pool.
3. financial product recommender system according to claim 2, which is characterized in that further include:
Normalized module carries out normalizing to the original daily earning rate of the financial product sample for utilizing preset formula
Change processing, obtains normalizing daily earning rate accordingly；
Sample value determining module, for using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate that the original daily earning rate, min indicate institute
The original daily earning rate of minimum in preset number of days is stated, max indicates the original daily earning rate of maximum in the preset number of days.
4. financial product recommender system according to claim 3, which is characterized in that
The quantitative value determining module, specifically for determining present count magnitude N using elbow method.
5. a kind of financial product recommended method characterized by comprising
Determine present count magnitude N；
According to the present count magnitude N, N number of financial product sample is selected from default financial product sample pool as N number of first
Beginning cluster centre；
Based on N number of initial cluster center, and ISODATA algorithm is utilized, to the institute in the default financial product sample pool
There is financial product sample to carry out cluster operation, obtains corresponding cluster result；
According to each final cluster centre in the cluster result, multiple financial product samples to be recommended are determined；
By preset communication interface by multiple financial product sample deliveries to be recommended to corresponding subscriber terminal equipment.
6. financial product recommended method according to claim 5, which is characterized in that before the determining present count magnitude N,
Further include:
The financial product in preset number of days is crawled from preset financial product database, obtains the default financial product sample
Pond.
7. financial product recommended method according to claim 6, which is characterized in that before the determining present count magnitude N,
Further include:
Using preset formula, the original daily earning rate of the financial product sample is normalized, is returned accordingly
One changes daily earning rate；
Using the normalization daily earning rate as the sample value of the financial product sample；
Wherein, the preset formula are as follows:
P_value*=(P_value-min)/max-min；
Wherein, P_value*Indicate that the normalization daily earning rate, P_value indicate that the original daily earning rate, min indicate institute
The original daily earning rate of minimum in preset number of days is stated, max indicates the original daily earning rate of maximum in the preset number of days.
8. financial product recommended method according to claim 7, which is characterized in that the determining present count magnitude N, packet
It includes:
Present count magnitude N is determined using elbow method.
9. a kind of financial product recommendation apparatus characterized by comprising
Memory, for saving computer program；
Processor, for executing the computer program, to realize as the described in any item financial products of claim 5 to 8 push away
Recommend method.
10. a kind of computer readable storage medium, which is characterized in that for saving computer program, wherein the computer journey
Such as claim 5 to 8 described in any item financial product recommended methods are realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811481623.XA CN109583777A (en) | 2018-12-05 | 2018-12-05 | A kind of financial product recommender system, method, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811481623.XA CN109583777A (en) | 2018-12-05 | 2018-12-05 | A kind of financial product recommender system, method, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109583777A true CN109583777A (en) | 2019-04-05 |
Family
ID=65927308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811481623.XA Pending CN109583777A (en) | 2018-12-05 | 2018-12-05 | A kind of financial product recommender system, method, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109583777A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110086874A (en) * | 2019-04-30 | 2019-08-02 | 清华大学 | A kind of Expressway Service user classification method, system, equipment and medium |
| CN110766208A (en) * | 2019-10-09 | 2020-02-07 | 中电科新型智慧城市研究院有限公司 | Government affair service demand prediction method based on social group behaviors |
| CN111832986A (en) * | 2020-06-05 | 2020-10-27 | 广东工业大学 | Product storage method and system, storage medium and computer equipment |
| CN117035948A (en) * | 2023-10-10 | 2023-11-10 | 山东唐和智能科技有限公司 | Task intelligent processing method and system based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440398A (en) * | 2013-07-10 | 2013-12-11 | 武汉大学 | Pattern recognition-based power grid branch importance estimation method |
| CN106897797A (en) * | 2017-02-23 | 2017-06-27 | 南京大学 | A kind of stock index tracking prediction method and system based on social network clustering |
| CN107909478A (en) * | 2017-11-27 | 2018-04-13 | 苏州点对点信息科技有限公司 | FOF mutual fund portfolio system and methods based on social network clustering and information gain entropy index |
| CN108009925A (en) * | 2017-11-27 | 2018-05-08 | 苏州点对点信息科技有限公司 | A kind of fund asset configuration system and method based on network clustering |
-
2018
- 2018-12-05 CN CN201811481623.XA patent/CN109583777A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440398A (en) * | 2013-07-10 | 2013-12-11 | 武汉大学 | Pattern recognition-based power grid branch importance estimation method |
| CN106897797A (en) * | 2017-02-23 | 2017-06-27 | 南京大学 | A kind of stock index tracking prediction method and system based on social network clustering |
| CN107909478A (en) * | 2017-11-27 | 2018-04-13 | 苏州点对点信息科技有限公司 | FOF mutual fund portfolio system and methods based on social network clustering and information gain entropy index |
| CN108009925A (en) * | 2017-11-27 | 2018-05-08 | 苏州点对点信息科技有限公司 | A kind of fund asset configuration system and method based on network clustering |
Non-Patent Citations (1)
| Title |
|---|
| 谢日敏 等: "高校信息系统使用率聚类分析", 高校《西安文理学院学报（自然科学版）》, vol. 19, no. 5, pages 41 - 43 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110086874A (en) * | 2019-04-30 | 2019-08-02 | 清华大学 | A kind of Expressway Service user classification method, system, equipment and medium |
| CN110766208A (en) * | 2019-10-09 | 2020-02-07 | 中电科新型智慧城市研究院有限公司 | Government affair service demand prediction method based on social group behaviors |
| CN111832986A (en) * | 2020-06-05 | 2020-10-27 | 广东工业大学 | Product storage method and system, storage medium and computer equipment |
| CN117035948A (en) * | 2023-10-10 | 2023-11-10 | 山东唐和智能科技有限公司 | Task intelligent processing method and system based on big data |
| CN117035948B (en) * | 2023-10-10 | 2024-01-09 | 山东唐和智能科技有限公司 | Task intelligent processing method and system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583777A (en) | A kind of financial product recommender system, method, equipment and medium | |
| Jasova et al. | Rollover risk and bank lending behavior: Evidence from unconventional central bank liquidity | |
| WO2011091054A1 (en) | Computing trading strategies for portfolio management and associated probability distributions for option pricing | |
| CN110264342A (en) | A kind of business audit method and device based on machine learning | |
| Lu et al. | A clustering-based portfolio strategy incorporating momentum effect and market trend prediction | |
| CN108428178A (en) | financing risk control method, device, equipment and storage medium | |
| Cai et al. | Gaussian weighting reversion strategy for accurate online portfolio selection | |
| Zhou et al. | Understanding the quintile portfolio | |
| Kašćelan et al. | Hybrid support vector machine rule extraction method for discovering the preferences of stock market investors: Evidence from Montenegro | |
| US20140279683A1 (en) | Performance Evaluation Of Mortgage Portfolios | |
| Brabazon et al. | Credit classification using grammatical evolution | |
| CN111292188A (en) | Intelligent integrated return test system for pairing transaction | |
| CN112767132B (en) | Data processing method and system | |
| CN107194814A (en) | The method and apparatus of recommended user | |
| Alaluf et al. | Reinforcement learning paycheck optimization for multivariate financial goals | |
| Wang et al. | The impact of RMB internationalization on the exchange rate linkages in China and ASEAN countries | |
| CN110033376A (en) | A kind of assets management method and device, a kind of calculating equipment and storage medium | |
| CN109544346A (en) | A kind of controllable type investment combination share-selecting method based on AP clustering algorithm | |
| Chio | A comparative study of the MACD-base trading strategies: evidence from the US stock market | |
| CN107305662A (en) | Recognize the method and device of violation account | |
| CN107092919A (en) | A kind of user's sample characteristics optimized treatment method and device | |
| CN111242720A (en) | Method and system for recommending suppliers for shopping providers | |
| CN109447814A (en) | Financial asset analysis method and device | |
| Jankovic et al. | Financial sector development and economic growth | |
| Papaioannou et al. | S&P500 Forecasting and trading using convolution analysis of major asset classes |
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 |