CN112667911A - Method for searching potential customers by using social software big data - Google Patents
Method for searching potential customers by using social software big data Download PDFInfo
- Publication number
- CN112667911A CN112667911A CN202110046743.2A CN202110046743A CN112667911A CN 112667911 A CN112667911 A CN 112667911A CN 202110046743 A CN202110046743 A CN 202110046743A CN 112667911 A CN112667911 A CN 112667911A
- Authority
- CN
- China
- Prior art keywords
- model
- screening
- data
- clients
- screening model
- 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 30
- 238000012216 screening Methods 0.000 claims abstract description 82
- 238000007637 random forest analysis Methods 0.000 claims description 16
- 238000012360 testing method Methods 0.000 claims description 16
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method for searching potential customers by utilizing social software big data, which can automatically search potential customers meeting conditions according to set screening conditions through a first screening model, further screen through a second screening model, and send customized information to the customers to screen the customers with higher success rate.
Description
Technical Field
The invention relates to a method for searching potential customers by utilizing social software big data.
Background
Developing new customers is one of the most important factors for business development. However, finding a new customer is a very difficult process that requires a significant investment of time and money, and still does not guarantee a smooth development of the new customer.
There are many payment software on the market that can provide information about potential customers based on several conditions (geographical location, industry, etc.) who can then try to contact them by phone, email, etc. Telemarketing or sending email is the traditional method of developing new customers, but is generally not effective and conversion rates are low. Because the transmission is not carried out aiming at the potential client really in need, most of the situations are that the opposite party does not reply or does not receive your call at all, and therefore a method for quickly and accurately screening the potential client in need by utilizing big data of social software is urgently needed at present.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a method for establishing social software big data to search potential customers.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a method for searching potential customers by utilizing big data of social software is characterized by comprising the following steps: the method comprises the following steps:
s1, creating a first screening model and a second screening model according to the information of the historical clients as data,
s2, setting screening conditions including education background, education field, current working company, working years, job title, whether any 3D or CAD software is used for the first screening model, automatically searching new clients meeting the screening conditions on the network according to the screening conditions,
s3, inputting the information of the new client searched by the first screening model into the second screening model and adding the industry and company scale as the screening condition of the second screening model to screen again,
s4, if the condition is met, the customized information is sent to contact the new client.
The creation of the first screening model and the second screening model comprises the following steps:
s1.1, collecting data, selecting a certain number of clients from historical clients and collecting information of the clients as data, wherein the clients comprise clients who have placed orders and clients who send price inquiry to me department but do not place orders at all, the information comprises education background, education field, current employment company, employment age, employment duty and whether any 3D or CAD software is used,
s1.2, cleaning data, discarding rows which can not copy data,
s1.3, analyzing data, searching correlation among various characteristics,
and S1.4, feature engineering, namely classifying partial features into variables and converting the variables into numerical values.
S1.5, creating a model, creating a plurality of random forest models,
s1.6, evaluating a model I, evaluating a random forest by checking a confusion matrix, a recall result and specificity, selecting the model I with the highest accuracy as a screening model I,
s1.7, performing a first model test, randomly selecting a certain number of customers from the social network site as a test, judging and screening the effect of the first model according to the success rate,
s1.8, creating a second screening model which is a random forest model identical to the first screening model, taking the output data of the first screening model as the input data of the second screening model and adding the scale of industries and companies as characteristics,
s1.9, evaluating the model II, detecting and screening the precision of the model II,
and S1.10, testing the model II, improving the precision of the screening model II through the super-parameter adjustment, and completing the establishment of the model.
The data analysis also includes analyzing all variables and creating chart analysis data, and building a machine learning model from the data to find correlations between various features.
The chart comprises a post list, a current employment company scale list and a study field list.
The invention has the beneficial effects that: the screening model I can automatically search potential customers meeting the conditions according to the set screening conditions, then the screening model II is further screened, and then customized information is sent to the customers to screen the customers with higher success rate.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is an analysis diagram of job types;
FIG. 2 is an analysis diagram of a type of academic calendar;
FIG. 3 is a company-scale analysis diagram;
fig. 4 is a flow chart of the present invention.
Detailed Description
Referring to fig. 1 to 4, the invention discloses a method for finding potential customers by using social software big data, which comprises the following steps:
s1, creating a first screening model and a second screening model according to the information of the historical clients as data,
s2, setting screening conditions for the first screening model, wherein the screening conditions include education background, education field, current working company, working years, working duties, and whether any 3D or CAD software is used, the first screening model automatically searches new clients meeting the screening conditions on the network according to the screening conditions, for example, the screening conditions can be one of the group of the family, the mechanical profession, the American group, the working period of more than three years, the management layer, the 3D or CAD software, thus the required clients can be preliminarily screened according to the conditions,
s3, inputting the information of the new client searched by the first screening model into the second screening model, and adding the industry and company scale as the screening condition of the second screening model to be screened again, wherein the screening condition is that the household appliance industry, the company earns more than ten million or the number of the company exceeds one hundred, and the like,
s4, if the condition is met, the customized information is sent to contact the new client.
The creation of the first screening model and the second screening model comprises the following steps:
s1.1, collecting data, selecting 2000 history clients and collecting information of the history clients as data, wherein the history clients comprise clients who have placed orders and clients who send inquiry prices to me department but do not place orders finally, the information comprises education background, education field, current employment company, employment age, employment duty and whether any 3D or CAD software is used, the information of the clients is mixed in a database, and then the information disclosed on social software is collected,
s1.2, cleaning data, because the collected data is chaotic, cleaning the data by using an interpolation method appropriately, discarding a plurality of lines which can not copy the data,
s1.3, analyzing data, in the process, analyzing the whole data, searching the correlation among various characteristics, determining which characteristics are important variables, determining the distribution of various data and target variables,
the analysis process comprises the steps of analyzing all variables and creating chart analysis data, and a machine learning model is built by adopting the data, wherein the data comprise different characteristics, so that more data are obtained through the charts, and the distribution of the target variables and the relation among the variables are determined. For example, it tells us how many of the customer's contacts who placed the order are engineers or purchases (as shown in FIG. 1, the abscissa represents the position and the ordinate represents the number of people corresponding to the position); how many clients 'contacts' scholars are masters or doctors (as shown in fig. 2, the abscissa represents the scholars, and the ordinate represents the number of people corresponding to the scholars); how large the number of customers is (as shown in fig. 3, the abscissa indicates the number range and the ordinate indicates the revenue corresponding to the number range), and so on. Then we analyze the data, in order to find out the characteristics, we check the correlation between the characteristics and the target variables, we use the built-in function of the imported variables, construct a basic model and draw a relevant chart, which can tell us which variables are most important, i arrange the variables in ascending order,
the objective variable may distinguish between customers who placed orders and customers who did not place orders. The distribution of this variable is about 49% to 51%. 49% of the customers placed orders, 51% of the people not placed orders,
s1.4, feature engineering, and classifying partial features into variables, wherein the variable formats are different. To use these variables, one-hot encoding techniques are used to convert them to numerical values,
s1.5, model creation, wherein a plurality of random forest models are created, a training set is needed for model creation firstly, and then a testing set is needed for model creation, the random forest models are trained through the training set and tested through the testing set, and the random forest models can be created by using a method of 60: 40, 70: 30 or 80: 20, preferably 70: 30, the data are enough for model test, and are randomly divided into two groups according to the total amount of the data and the ratio, the data set is randomly divided, and the two groups contain various variables. After dividing the data by a ratio of 70-30, we created a random forest model.
The random forest model distributes data in a random manner, and builds a plurality of decision trees, and then takes the average value of all the decision trees in a democratic manner. The random forest has less required computing power, is easy to deploy and can completely meet the requirements of people.
S1.6, evaluating a model I, evaluating a random forest by checking a confusion matrix, a recall result and specificity, selecting the model I with the highest accuracy as a screening model I,
s1.7, performing model I test, namely randomly selecting a certain number of customers from a social network site as a test, judging and screening the effect of the model I according to the success rate, and performing the test by using the original information of about 500 people in the Ying website. After the screening model I runs the data, 140 potential customers are screened for us, then the customers send invitations for establishing contact in the captain, the acceptance rate of the invitations screened manually is about 20% before, but now reaches about 80%, so that the model is seen to instantly improve the conversion rate and help the customers to find out more suitable potential customers from the original data.
S1.8, creating a second screening model, wherein the second screening model is a random forest model identical to the first screening model, the output data of the first screening model is used as the input data of the second screening model, the scale of industries and companies is added as the characteristics, and the overall accuracy of the first screening model is about 82%. But the recall is (1, 0-62, 98). This means that the model achieves 98% accuracy when screening out unsuitable people, but only 62% accuracy when selecting matching customers. On the basis of the establishment of a second screening model, the second screening model can screen out more persons unlikely to place orders to the second screening model, in the second screening model, more specific characteristics are needed to separate potential customers from non-customers, therefore, the second screening model decides to use the characteristics of the industry and the company scale, as a manufacturing enterprise, some industries and company scales are very suitable for the second screening model, therefore, the purpose of the second screening model is to filter out other persons which are not suitable again from the output of the first screening model,
s1.9, evaluating the model II, detecting the precision of the screened model II, evaluating the precision of the screened model II by checking a confusion matrix, a recall result and specificity, wherein the precision of the screened model II reaches 85 percent and is obviously improved compared with the first model,
and S1.10, testing the model II, improving the precision of the screening model II through the super-parameter adjustment, completing the establishment of the model, and finally improving the accuracy of the model to 89%.
In summary, the artificial intelligence customer source filter built according to the principle of the method is sales efficiency software, can be matched with any type of market development tools such as Ying, attention information (American B2B marketing intelligent service company), Networks (Germany business social network site) and the like, and remarkably reduces junk mails generated by company sales and market departments through more accurate customer source filtering.
The method for finding potential customers by using social software big data provided by the embodiment of the invention is described in detail above, a specific example is applied in the method to explain the principle and the implementation of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (5)
1. A method for searching potential customers by utilizing big data of social software is characterized by comprising the following steps: the method comprises the following steps:
s1, creating a first screening model and a second screening model according to the information of the historical clients as data,
s2, setting screening conditions including education background, education field, current working company, working years, job title, whether any 3D or CAD software is used for the first screening model, automatically searching new clients meeting the screening conditions on the network according to the screening conditions,
s3, inputting the information of the new client searched by the first screening model into the second screening model and adding the industry and company scale as the screening condition of the second screening model to screen again,
s4, if the condition is met, the customized information is sent to contact the new client.
2. The method for finding potential customers by utilizing big data of social software according to claim 1, wherein the method comprises the following steps: the creation of the first screening model and the second screening model comprises the following steps:
s1.1, collecting data, selecting a certain number of clients from historical clients and collecting information of the clients as data, wherein the clients comprise clients who have placed orders and clients who send price inquiry to me department but do not place orders at all, the information comprises education background, education field, current employment company, employment age, employment duty and whether any 3D or CAD software is used,
s1.2, cleaning data, discarding rows which can not copy data,
s1.3, analyzing data, searching correlation among various characteristics,
and S1.4, feature engineering, namely classifying partial features into variables and converting the variables into numerical values.
S1.5, creating a model, creating a plurality of random forest models,
s1.6, evaluating a model I, evaluating a random forest by checking a confusion matrix, a recall result and specificity, selecting the model I with the highest accuracy as a screening model I,
s1.7, performing a first model test, randomly selecting a certain number of customers from the social network site as a test, judging and screening the effect of the first model according to the success rate,
s1.8, creating a second screening model which is a random forest model identical to the first screening model, taking the output data of the first screening model as the input data of the second screening model and adding the scale of industries and companies as characteristics,
s1.9, evaluating the model II, detecting and screening the precision of the model II,
and S1.10, testing the model II, improving the precision of the screening model II through the super-parameter adjustment, and completing the establishment of the model.
3. The method for finding potential customers by utilizing big data of social software according to claim 2, wherein the method comprises the following steps: the data analysis also includes analyzing all variables and creating chart analysis data, and building a machine learning model from the data to find correlations between various features.
4. The method for finding potential customers by utilizing big data of social software according to claim 1, wherein the method comprises the following steps: the chart comprises a post list, a current employment company scale list and a study field list.
5. The method for finding potential customers by utilizing big data of social software according to claim 1, wherein the method comprises the following steps: the model creation in S1.5 further includes creating a training set and a test set, training the random forest model through the training set, and testing the random forest model through the test set.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110046743.2A CN112667911A (en) | 2021-01-14 | 2021-01-14 | Method for searching potential customers by using social software big data |
PCT/CN2021/073532 WO2022151524A1 (en) | 2021-01-14 | 2021-01-25 | Method for seeking potential customers by using social software big data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110046743.2A CN112667911A (en) | 2021-01-14 | 2021-01-14 | Method for searching potential customers by using social software big data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112667911A true CN112667911A (en) | 2021-04-16 |
Family
ID=75415145
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110046743.2A Pending CN112667911A (en) | 2021-01-14 | 2021-01-14 | Method for searching potential customers by using social software big data |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN112667911A (en) |
WO (1) | WO2022151524A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177151A (en) * | 2021-05-28 | 2021-07-27 | 中山世达模型制造有限公司 | Potential customer screening method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
CN106228389A (en) * | 2016-07-14 | 2016-12-14 | 武汉斗鱼网络科技有限公司 | Network potential usage mining method and system based on random forests algorithm |
CN106779827A (en) * | 2016-12-02 | 2017-05-31 | 上海晶樵网络信息技术有限公司 | A kind of Internet user's behavior collection and the big data method of analysis detection |
CN108256052A (en) * | 2018-01-15 | 2018-07-06 | 成都初联创智软件有限公司 | Automobile industry potential customers' recognition methods based on tri-training |
US20180225685A1 (en) * | 2017-02-07 | 2018-08-09 | Linkedin Corporation | Identifying impending user-competitor relationships on an online social networking system |
US10289738B1 (en) * | 2013-12-23 | 2019-05-14 | Massachusetts Mutual Life Insurance Company | System and method for identifying potential clients from aggregate sources |
CN110009416A (en) * | 2019-04-02 | 2019-07-12 | 安徽筋斗云机器人科技股份有限公司 | A kind of system based on big data cleaning and AI precision marketing |
CN110222272A (en) * | 2019-04-18 | 2019-09-10 | 广东工业大学 | A kind of potential customers excavate and recommended method |
-
2021
- 2021-01-14 CN CN202110046743.2A patent/CN112667911A/en active Pending
- 2021-01-25 WO PCT/CN2021/073532 patent/WO2022151524A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10289738B1 (en) * | 2013-12-23 | 2019-05-14 | Massachusetts Mutual Life Insurance Company | System and method for identifying potential clients from aggregate sources |
CN105488697A (en) * | 2015-12-09 | 2016-04-13 | 焦点科技股份有限公司 | Potential customer mining method based on customer behavior characteristics |
CN106228389A (en) * | 2016-07-14 | 2016-12-14 | 武汉斗鱼网络科技有限公司 | Network potential usage mining method and system based on random forests algorithm |
CN106779827A (en) * | 2016-12-02 | 2017-05-31 | 上海晶樵网络信息技术有限公司 | A kind of Internet user's behavior collection and the big data method of analysis detection |
US20180225685A1 (en) * | 2017-02-07 | 2018-08-09 | Linkedin Corporation | Identifying impending user-competitor relationships on an online social networking system |
CN108256052A (en) * | 2018-01-15 | 2018-07-06 | 成都初联创智软件有限公司 | Automobile industry potential customers' recognition methods based on tri-training |
CN110009416A (en) * | 2019-04-02 | 2019-07-12 | 安徽筋斗云机器人科技股份有限公司 | A kind of system based on big data cleaning and AI precision marketing |
CN110222272A (en) * | 2019-04-18 | 2019-09-10 | 广东工业大学 | A kind of potential customers excavate and recommended method |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113177151A (en) * | 2021-05-28 | 2021-07-27 | 中山世达模型制造有限公司 | Potential customer screening method |
WO2022246923A1 (en) * | 2021-05-28 | 2022-12-01 | 中山世达模型制造有限公司 | Method for screening potential customer |
Also Published As
Publication number | Publication date |
---|---|
WO2022151524A1 (en) | 2022-07-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110110881B (en) | Power customer demand prediction analysis method and system | |
Popovic et al. | Quantitative indicators for social sustainability assessment of society and product responsibility aspects in supply chains | |
Chatterjee et al. | Supplier selection in Telecom supply chain management: a Fuzzy-Rasch based COPRAS-G method | |
CN112668859A (en) | Big data based customer risk rating method, device, equipment and storage medium | |
CN112907305B (en) | Customer full-period management system based on big data analysis | |
CN104321794A (en) | A system and method using multi-dimensional rating to determine an entity's future commercial viability | |
CN113051291A (en) | Work order information processing method, device, equipment and storage medium | |
CN111639690A (en) | Fraud analysis method, system, medium, and apparatus based on relational graph learning | |
CN105740434A (en) | Network information scoring method and device | |
CN115759640A (en) | Public service information processing system and method for smart city | |
CN117217634B (en) | Enterprise cooperation community discovery method based on complex network | |
CN112950359B (en) | User identification method and device | |
CN112667911A (en) | Method for searching potential customers by using social software big data | |
Gupta et al. | Construction and demolition waste causative factors in building projects: survey of the Indian construction industry experiences | |
CN106897198A (en) | A kind of processing method and processing device of daily record data | |
CN113283806A (en) | Enterprise information evaluation method and device, computer equipment and storage medium | |
Triantis | Fuzzy non-radial data envelopment analysis (DEA) measures of technical efficiency in support of an integrated performance measurement system | |
CN113379432B (en) | Sales system customer matching method based on machine learning | |
CN115147091A (en) | Intelligent salary query method and system | |
JP2022120816A (en) | New product sales forecast method | |
CN114511250A (en) | Enterprise external migration risk early warning method and system based on machine learning | |
JP2017200079A (en) | Computing for outputting distrust index of user by acquiring telephone number of user declared from portable terminal through internet | |
US8589209B2 (en) | System and method for assessing viability and marketability of assets | |
CN112528143A (en) | Intelligent pushing method and system for price inquiring order | |
CN116342300B (en) | Method, device and equipment for analyzing characteristics of insurance claim settlement personnel |
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 |