CN111563628A - Real estate customer transaction time prediction method, device and storage medium - Google Patents
Real estate customer transaction time prediction method, device and storage medium Download PDFInfo
- Publication number
- CN111563628A CN111563628A CN202010388361.3A CN202010388361A CN111563628A CN 111563628 A CN111563628 A CN 111563628A CN 202010388361 A CN202010388361 A CN 202010388361A CN 111563628 A CN111563628 A CN 111563628A
- Authority
- CN
- China
- Prior art keywords
- customer
- client
- transaction
- predicted
- type
- 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 claims abstract description 37
- 238000013145 classification model Methods 0.000 claims abstract description 44
- 238000003064 k means clustering Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 23
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims description 15
- 230000003203 everyday effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000004904 shortening Methods 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/16—Real estate
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a real estate customer transaction time prediction method, a real estate customer transaction time prediction device and a storage medium, wherein model input information of a customer to be predicted is obtained; inputting the data into a KNN classification model, and determining a client type to which a client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering the model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types; determining a transaction time interval of the target client type according to the transaction clients belonging to the target client type in the clustering result; predicting the transaction time of the client to be predicted according to the transaction time interval of the target client type; the method is beneficial to targeted and accurate marketing, improves the conversion of the transaction rate and reduces the marketing cost.
Description
Technical Field
The invention relates to the technical field of real estate Internet, in particular to a real estate customer transaction time prediction method, a real estate customer transaction time prediction device and a storage medium.
Background
With the rise of the internet, shopping methods such as on-line house selection are gradually merged into real estate transactions, but because the full right of the initiative of on-line house selection is unidirectionally controlled by house purchasers, how to find time to carry out accurate marketing on customers is a problem to be solved by current real estate manufacturers. Each customer has a time period with the strongest purchasing intention, and the time period is the time period which is easiest to deal with, so that the customers are recommended and marketed, and the purpose of reducing marketing cost and promoting the deal rate is achieved.
Disclosure of Invention
The invention provides a real estate customer transaction time prediction method, a device and a storage medium, which mainly solve the technical problems that: how to predict the time period in which the real estate non-transaction client is most likely to make a transaction is beneficial to carrying out targeted and accurate marketing, improving the conversion of the transaction rate and reducing the marketing cost.
In order to solve the technical problem, the invention provides a real estate customer transaction time prediction method, which comprises the following steps:
obtaining model input information of a customer to be predicted, wherein the model input information is obtained by processing on-line behavior data and/or customer file information of the customer;
inputting the prediction data into a KNN classification model, and determining a client type to which the client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types;
determining a transaction time interval of the target customer type according to the transaction customers belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time;
and predicting the transaction time of the customer to be predicted according to the transaction time interval of the target customer type.
Optionally, the KNN classification model is obtained by performing a training test using a K-means clustering result, and includes:
and randomly selecting 70% and 30% of the committed clients in the K-means clusters as training data and test data respectively to determine the K value of the KNN classification model so as to obtain the KNN classification model.
Optionally, the client online behavior data includes: browsing times, the residence time of the most browsed building, the total number of types of households browsed, the total time of paying attention to the building and browsing frequency every day;
the customer file information comprises: client age, house buying application, house buying category and house watching duration; wherein the house buying purpose comprises self-residence and investment; the house purchasing category comprises just-needed, improved and invested.
Optionally, the inputting to the KNN classification model, and the determining the customer type to which the customer to be predicted belongs includes:
calculating the distance between the model input information of the customer to be predicted and each training data in the KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
Optionally, the determining, according to the committed clients belonging to the target client type in the clustering result, the commit time interval of the target client type includes:
and acquiring the respective transaction time intervals of the transaction clients belonging to the target client type in the clustering result, and screening the transaction time interval with the highest occurrence frequency as the transaction time interval of the target client type.
Optionally, before inputting the model input information of the customer to be predicted into the KNN classification model, the method further includes: and predicting the transaction probability of the client to be predicted by using a LightGBM model, comparing the transaction probability with a set threshold value, and judging that the transaction probability of the client to be predicted reaches the set threshold value.
The invention also provides a real estate customer transaction time prediction device, comprising:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring model input information of a client to be predicted, and the model input information is obtained by processing on-line behavior data and/or client file information of the client;
the classification module is used for inputting the classification model into the KNN classification model, determining the client type of the client to be predicted as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types;
the processing module is used for determining the transaction time interval of the target customer type according to the transaction clients belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time; and predicting the transaction time of the customer to be predicted according to the transaction time interval of the target customer type.
Optionally, the classification module is configured to calculate a distance between the model input information of the customer to be predicted and each training data in the KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
Optionally, the processing module is configured to obtain respective deal time intervals of the committed clients belonging to the target client type in the clustering result, and filter a deal time interval with the highest occurrence frequency as a deal time interval of the target client type.
The present invention also provides a storage medium storing one or more programs executable by one or more processors to implement the steps of the real estate customer deal time prediction method described above.
The invention has the beneficial effects that:
according to the real estate customer transaction time prediction method, the real estate customer transaction time prediction device and the storage medium, model input information of a customer to be predicted is obtained by obtaining the model input information, wherein the model input information is obtained by processing on-line behavior data and/or customer file information of the customer; inputting the data into a KNN classification model, and determining a client type to which a client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering the model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types; determining a transaction time interval of the target client type according to the transaction clients belonging to the target client type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time; predicting the transaction time of the client to be predicted according to the transaction time interval of the target client type; the method is beneficial to targeted and accurate marketing, improves the conversion of the transaction rate and reduces the marketing cost.
Drawings
FIG. 1 is a flow chart of a real estate customer deal time prediction method according to an embodiment of the invention;
fig. 2 is a schematic structural diagram of a real estate customer transaction time prediction device according to a second embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following detailed description and accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The first embodiment is as follows:
in order to solve the problem that in the traditional real estate recommending and marketing process, a local producer cannot accurately know potential customers who do not make a deal in which time period the deal intention is the strongest, so that the marketing can be performed blindly, the customers are usually uninteresting when contacting the customers, or the house buying and other conditions are not considered in the near period, so that the marketing is performed blindly in the time period with weaker client deal intention, the deal cannot be promoted, the marketing investment of the local producer is improved, and meanwhile, the bad experience is brought to the customers possibly due to the disturbance of the normal work and life of the customers. In conclusion, how to predict the time period in which the purchasing intention of potential non-bargained customers is stronger and further carry out accurate marketing is a problem which is continuously solved by the local producer, so that the conversion of the bargaining rate is improved, and the marketing cost is reduced.
The embodiment provides a real estate customer transaction time prediction method, which clusters customers according to browsing behavior data and customer profile data of committed customers (clustering customers with similar browsing behavior and customer profile data of committed customers into one class), classifies non-committed customers according to a clustering result, matches transaction time intervals of committed customers under the class according to a classification result, predicts transaction time of the non-committed customers, and distributes business consultants to actively contact the customers around the time point, so that the purposes of shortening transaction time and reducing marketing cost are achieved.
Referring to fig. 1, the method for predicting the transaction time of a real estate customer mainly comprises the following steps:
s101, obtaining model input information of a customer to be predicted, wherein the model input information is obtained by processing on-line behavior data and/or customer file information of the customer.
The client to be predicted can be a non-transaction client of the project or any client with a house purchasing requirement, and can be flexibly set according to the actual requirement. In order to ensure the effectiveness of the prediction result and avoid predicting unnecessary clients, for example, it does not make a substantial sense to predict the transaction time of a client who cannot make a transaction, in other embodiments of the present invention, before predicting the client to be predicted, the LightGBM model is used to predict the transaction probability of the client to be predicted, and the transaction probability is compared with the set threshold, and it is determined that the transaction probability of the client to be predicted reaches the set threshold. And screening out potential customers with the transaction probability being greater than or equal to a set threshold value for prediction.
The LightGBM prediction model can be obtained by selecting 30000 transaction client records as positive samples, selecting the same non-transaction clients as negative samples according to downsampling, dividing the data into training data and testing data according to the ratio of 7:3, and training and testing the model. Inputting a piece of non-deal customer data, and outputting a deal probability value of the customer by the model. If the transaction probability is more than 60%, the client can be used as the client to be predicted to predict the transaction time interval.
In predicting the customer deal probability, the following data can be selected: customer ID, visit building ID, visit days, total visit page number, total browse duration, total browse times, visit building number, visit house type, visit night, average visit daily duration, average click times per day, average visit page number per day, maximum click times per day, maximum browse time per day, number of days visited before and after, use number of housing loan calculators, whether to meet or not, and the like.
S102, inputting the data into the KNN classification model, and determining a client type to which a client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering the model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types.
The model input information of the customer to be predicted is mainly based on the online behavior data and the customer file information of the customer. Before inputting the KNN classification model, the KNN classification model needs to be preprocessed to meet the model input requirement.
Wherein, the client online behavior data comprises: browsing times, the residence time of the most browsed building, the total number of types of households browsed, the total time of paying attention to the building and browsing frequency every day;
the customer file information includes: client age, house buying application, house buying category and house watching duration; wherein the house buying purpose comprises self-residence and investment; the house purchasing category comprises just-needed, improved and invested.
Preprocessing the model input information of the client to be predicted, including performing data normalization on the model input information using z-scores: first, an arithmetic mean (mathematical expectation) x of each index is obtainediAnd standard deviation si(ii) a Then, each index is subjected to standardization treatment: z is a radical ofij＝(xij-xi)/si(ii) a Wherein: z is a radical ofijThe normalized index value is obtained; x is the number ofijIs an actual index value; and finally, adjusting the negative sign before the inverse index to be positive.
The KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, behavior data and customer profile data of a committed customer are taken, and an elbow method in clustering analysis is utilized to determine a hyper-parameter K so as to avoid local optimization; and judging the category of the client, carrying out k-means clustering, and marking a category label on the client.
Specifically, (1) K points are selected as an initial centroid; (2) assigning each point to the nearest centroid, forming K clusters; (3) recalculating the centroid of each cluster; repeating the steps (2) to (3) until the cluster is not changed or the maximum iteration number is reached; aggregating customers as Q1、Q2、......、QK。
And randomly selecting 70% and 30% of committed clients in the K-means clusters as training data and testing data respectively, and calculating the accuracy of the corresponding model under different K values by using a recursion relation to determine the optimal K value of the KNN classification model so as to obtain the KNN classification model. It should be noted that the value of the hyper-parameter K in the K-means clustering model is not necessarily the same as the value of the number K of the adjacent points of the KNN classification model, and the two are not necessarily related.
Inputting the model input information of the customer to be predicted into the KNN classification model, and calculating the distance between the model input information of the customer to be predicted and each training data in the KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
For example, based on the model input information of the customer to be predicted, the model input information is input into the KNN classification model, the nearest 5 points are obtained and are respectively the committed customers A, B, C, D, E, and the assumption is that the customer A belongs to Q according to the clustering result1Customer B belongs to Q2Customer C belongs to Q1Client D belongs to Q3Client E belongs to Q4Then the most frequent customer type is Q1Thereby determining that the client type to be predicted belongs to is Q1。
S103, determining a transaction time interval of the target customer type according to the transaction customers belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time.
Specifically, the respective transaction time intervals of the already-delivered customers belonging to the target customer type in the clustering result are obtained, and the transaction time interval with the highest occurrence frequency is screened out and used as the transaction time interval of the target customer type.
For example, after determining that the type of the client to be predicted is Q1Then, according to the clustering result, obtaining the clustering result belonging to Q1The transaction time interval of all the transaction clients is assumed to be Q1The number of people with 45-day transaction time intervals is the largest among all the transaction clients, namely the occurrence frequency is the highest, so that the Q is determined1The trade time interval for type of customer is 45 days.
And S104, predicting the transaction time of the client to be predicted according to the transaction time interval of the target client type.
When the transaction time interval with the highest occurrence frequency of the target client type is obtained, the fact that the sum probability of the clients in the time period is the highest and the transaction intention is the strongest is shown, the client to be predicted belongs to the target client type, and then the transaction time interval of the target client type is used as the transaction time interval of the client to be predicted; or setting a time range on the basis of the transaction time interval of the target client type as the transaction time interval of the client to be predicted. For example, the transaction time interval of the target customer type is 45 days, the set time range is ± 5 days, that is, the transaction time interval of the customer to be predicted is 40-50 days. Furthermore, the business time interval of the client to be predicted can be sent to the corresponding business counselor, and the business counselor is informed to actively recommend marketing to the client when the business time interval is reached.
An effective means is provided for predicting the transaction time of the non-transaction client; meanwhile, before the deal time is predicted, the deal probability of the client is predicted through the LightGBM model, and the non-deal client with high deal probability is screened out to serve as the client to be predicted, so that the accuracy of the model can be greatly improved. Through actual data testing, the accuracy of the prediction scheme of the client bargain time is finally measured to be up to 72%, reliable technical support is provided for accurate marketing of local producers, the improvement of bargain conversion is facilitated, and the marketing cost is reduced.
The real estate customer transaction time prediction method provided by the invention obtains model input information of a customer to be predicted, wherein the model input information is obtained by processing on-line behavior data and/or customer file information of the customer; inputting the data into a KNN classification model, and determining a client type to which a client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering the model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types; determining a transaction time interval of the target client type according to the transaction clients belonging to the target client type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time; predicting the transaction time of the client to be predicted according to the transaction time interval of the target client type; the method is beneficial to targeted and accurate marketing, improves the conversion of the transaction rate and reduces the marketing cost.
Example two:
in this embodiment, on the basis of the first embodiment, a real estate customer transaction time prediction apparatus is provided for implementing at least part of the steps of the real estate customer transaction time prediction method, please refer to fig. 2, the apparatus mainly includes an obtaining module 21, a classifying module 22 and a processing module 23, wherein:
the obtaining module 21 is configured to obtain model input information of the customer to be predicted, where the model input information is obtained by processing online behavior data and/or customer profile information of the customer.
The classification module 22 is used for inputting the classification model into the KNN, and determining the client type of the client to be predicted as the target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering the model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types.
The processing module 23 is configured to determine a transaction time interval of the target customer type according to the committed customers belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time; and predicting the transaction time of the client to be predicted according to the transaction time interval of the target client type.
Further, the classification module 22 is configured to calculate a distance between the model input information of the customer to be predicted and each training data in the KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
Further, the processing module 23 is configured to obtain respective deal time intervals of the already-dealt customers belonging to the target customer type in the clustering result, and filter a deal time interval with the highest occurrence frequency as a deal time interval of the target customer type.
The present embodiments also provide a storage medium having one or more programs stored thereon that are executable by one or more processors to perform the steps of the real estate customer deal time prediction method described above.
It will be apparent to those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be centralized on a single computing device or distributed across a network of computing devices, and optionally they may be implemented in program code executable by a computing device, such that they may be stored on a computer storage medium (ROM/RAM, magnetic disks, optical disks) and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The foregoing is a more detailed description of the present invention that is presented in conjunction with specific embodiments, and the practice of the invention is not to be considered limited to those descriptions. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (10)
1. A real estate customer transaction time prediction method is characterized by comprising the following steps:
obtaining model input information of a customer to be predicted, wherein the model input information is obtained by processing on-line behavior data and/or customer file information of the customer;
inputting the prediction data into a KNN classification model, and determining a client type to which the client to be predicted belongs as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types;
determining a transaction time interval of the target customer type according to the transaction customers belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time;
and predicting the transaction time of the customer to be predicted according to the transaction time interval of the target customer type.
2. A real estate customer deal time prediction method as claimed in claim 1 wherein the KNN classification model is derived from a training test using K-means clustering results comprising:
and randomly selecting 70% and 30% of the committed clients in the K-means clusters as training data and test data respectively to determine the K value of the KNN classification model so as to obtain the KNN classification model.
3. A real estate customer deal time prediction method as claimed in claim 1 wherein the customer online behavior data comprises: browsing times, the residence time of the most browsed building, the total number of types of households browsed, the total time of paying attention to the building and browsing frequency every day;
the customer file information comprises: client age, house buying application, house buying category and house watching duration; wherein the house buying purpose comprises self-residence and investment; the house purchasing category comprises just-needed, improved and invested.
4. A real estate customer deal time prediction method as claimed in claim 2 wherein said inputting into a KNN classification model, determining the customer type to which the customer to be predicted belongs comprises:
calculating the distance between the model input information of the customer to be predicted and each training data in the KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
5. A real estate customer deal time prediction method as claimed in claim 1 wherein the determining a deal time interval for the target customer type based on the committed customers belonging to the target customer type in the clustering result comprises:
and acquiring the respective transaction time intervals of the transaction clients belonging to the target client type in the clustering result, and screening the transaction time interval with the highest occurrence frequency as the transaction time interval of the target client type.
6. A real estate customer deal time prediction method as claimed in claim 1 further comprising, prior to inputting the model input information for the customer to be predicted into the KNN classification model: and predicting the transaction probability of the client to be predicted by using a LightGBM model, comparing the transaction probability with a set threshold value, and judging that the transaction probability of the client to be predicted reaches the set threshold value.
7. A real estate customer deal time prediction apparatus, comprising:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring model input information of a client to be predicted, and the model input information is obtained by processing on-line behavior data and/or client file information of the client;
the classification module is used for inputting the classification model into the KNN classification model, determining the client type of the client to be predicted as a target client type; the KNN classification model is obtained by utilizing a K-means clustering result to carry out training test, wherein the K-means clustering process is used for clustering model input information of a plurality of already-handed customers, and the plurality of already-handed customers are clustered into a plurality of customer types;
the processing module is used for determining the transaction time interval of the target customer type according to the transaction clients belonging to the target customer type in the clustering result; the transaction time interval is the time interval between the earliest generation time of the model input information and the transaction time; and predicting the transaction time of the customer to be predicted according to the transaction time interval of the target customer type.
8. The real estate customer deal time prediction device of claim 7 wherein the classification module is configured to calculate distances between model input information of the customer to be predicted and training data in a KNN classification model; and selecting K points with the closest distance, and determining the client type with the highest occurrence frequency according to the client types to which the K points respectively belong to serve as the client type to which the client to be predicted belongs.
9. The real estate customer deal time prediction device of claim 8 wherein the processing module is configured to obtain the respective deal time intervals of the deal customers belonging to the target customer type in the clustering result, and filter the deal time interval with the highest frequency of occurrence as the deal time interval of the target customer type.
10. A storage medium storing one or more programs, the one or more programs being executable by one or more processors to perform the steps of the real estate client deal time prediction method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010388361.3A CN111563628A (en) | 2020-05-09 | 2020-05-09 | Real estate customer transaction time prediction method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010388361.3A CN111563628A (en) | 2020-05-09 | 2020-05-09 | Real estate customer transaction time prediction method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111563628A true CN111563628A (en) | 2020-08-21 |
Family
ID=72074674
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010388361.3A Pending CN111563628A (en) | 2020-05-09 | 2020-05-09 | Real estate customer transaction time prediction method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111563628A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112200220A (en) * | 2020-09-18 | 2021-01-08 | 中国航空无线电电子研究所 | Health monitoring method for airplane airborne equipment based on data induction |
| CN112561571A (en) * | 2020-12-07 | 2021-03-26 | 深圳市思为软件技术有限公司 | House purchasing intention evaluation method and related equipment |
| CN112950359A (en) * | 2021-03-30 | 2021-06-11 | 建信金融科技有限责任公司 | User identification method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230029A (en) * | 2017-12-29 | 2018-06-29 | 西南大学 | Client trading behavior analysis method |
| CN108898413A (en) * | 2018-05-14 | 2018-11-27 | 链家网（北京）科技有限公司 | A kind of traveller's management method and device |
| CN108921342A (en) * | 2018-06-26 | 2018-11-30 | 圆通速递有限公司 | A kind of logistics customer churn prediction method, medium and system |
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
| CN109615128A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | Real estate client's conclusion of the business probability forecasting method, device and server |
| CN109615426A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | A kind of marketing method based on Customer clustering, system |
| CN109785000A (en) * | 2019-01-16 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Customer resources distribution method, device, storage medium and terminal |
| CN110415002A (en) * | 2019-07-31 | 2019-11-05 | 中国工商银行股份有限公司 | Customer behavior prediction method and system |
-
2020
- 2020-05-09 CN CN202010388361.3A patent/CN111563628A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230029A (en) * | 2017-12-29 | 2018-06-29 | 西南大学 | Client trading behavior analysis method |
| CN108898413A (en) * | 2018-05-14 | 2018-11-27 | 链家网（北京）科技有限公司 | A kind of traveller's management method and device |
| CN108921342A (en) * | 2018-06-26 | 2018-11-30 | 圆通速递有限公司 | A kind of logistics customer churn prediction method, medium and system |
| CN109583651A (en) * | 2018-12-03 | 2019-04-05 | 焦点科技股份有限公司 | A kind of method and apparatus for insuring electric business platform user attrition prediction |
| CN109615128A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | Real estate client's conclusion of the business probability forecasting method, device and server |
| CN109615426A (en) * | 2018-12-05 | 2019-04-12 | 重庆锐云科技有限公司 | A kind of marketing method based on Customer clustering, system |
| CN109785000A (en) * | 2019-01-16 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Customer resources distribution method, device, storage medium and terminal |
| CN110415002A (en) * | 2019-07-31 | 2019-11-05 | 中国工商银行股份有限公司 | Customer behavior prediction method and system |
Non-Patent Citations (1)
| Title |
|---|
| 周爱国: ""基于K-means和K近邻的DPF设备故障分类算法"", 《内燃机与配件》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112200220A (en) * | 2020-09-18 | 2021-01-08 | 中国航空无线电电子研究所 | Health monitoring method for airplane airborne equipment based on data induction |
| CN112561571A (en) * | 2020-12-07 | 2021-03-26 | 深圳市思为软件技术有限公司 | House purchasing intention evaluation method and related equipment |
| CN112950359A (en) * | 2021-03-30 | 2021-06-11 | 建信金融科技有限责任公司 | User identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11586880B2 (en) | System and method for multi-horizon time series forecasting with dynamic temporal context learning | |
| Ribeiro et al. | Transfer learning with seasonal and trend adjustment for cross-building energy forecasting | |
| US20220358528A1 (en) | Methods and apparatus for self-adaptive time series forecasting engine | |
| US10348550B2 (en) | Method and system for processing network media information | |
| AU2019426881A1 (en) | Method and system of dynamic model selection for time series forecasting | |
| WO2021174944A1 (en) | Message push method based on target activity, and related device | |
| CN111563628A (en) | Real estate customer transaction time prediction method, device and storage medium | |
| CN109615129B (en) | Real estate customer transaction probability prediction method, server and computer storage medium | |
| US20210103858A1 (en) | Method and system for model auto-selection using an ensemble of machine learning models | |
| Ford et al. | A real-time self-adaptive classifier for identifying suspicious bidders in online auctions | |
| CN111814910B (en) | Abnormality detection method, abnormality detection device, electronic device, and storage medium | |
| US20230108469A1 (en) | Systems and methods for generating dynamic conversational responses using cluster-level collaborative filtering matrices | |
| CN111062564A (en) | Method for calculating power customer appeal sensitive value | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN113450141B (en) | Intelligent prediction method and device based on electricity sales quantity characteristics of large power customer group | |
| CN114584601A (en) | User loss identification and intervention method, system, terminal and medium | |
| CN111582394B (en) | Group assessment method, device, equipment and medium | |
| CN116523301A (en) | System for predicting risk rating based on big data of electronic commerce | |
| CN114520773B (en) | Service request response method, device, server and storage medium | |
| CN111325372A (en) | Method for establishing prediction model, prediction method, device, medium and equipment | |
| US20230075453A1 (en) | Generating machine learning based models for time series forecasting | |
| Overby et al. | How reduced search costs and the distribution of bidder participation affect auction prices | |
| Asghari et al. | Selecting and prioritizing the electricity customers for participating in demand response programs | |
| WO2021077227A1 (en) | Method and system for generating aspects associated with a future event for a subject | |
| Zhuang et al. | DyS-IENN: a novel multiclass imbalanced learning method for early warning of tardiness in rocket final assembly process |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200821 |