[0032] Example one:
[0033] In order to achieve bridging between buyers and sellers and promote transactions, we will accurately recommend customers with strong purchasing needs to technology transfer intentions to corresponding sellers, improve the efficiency of patent transactions, and make buyers better, more convenient and more efficient to buy at the right price , Patent whose technology meets their own needs, this embodiment provides a method for accurately recommending technology transfer intentions to customers, which is mainly applied to a patent transaction platform to promote rapid transactions between buyers and sellers of patents. Calculate the recommendation index to be recommended by analyzing the activity of the client to be recommended (equivalent to the buyer) on the patent trading platform, the patent information of the client to be recommended, and the basic information of the company; The higher the value, the stronger the purchase intention of the customer to be recommended, and the more likely the transaction will be achieved; on the contrary, the weaker the purchase intention of the customer to be recommended and the more difficult the transaction will be. Based on this, the customers who meet the purchase intention conditions are recommended to the corresponding sellers, so that the sellers directly contact actively, thereby promoting the transaction.
[0034] See figure 1 The method for accurately recommending technology transfer intention customers provided in this embodiment mainly includes the following steps:
[0035] S101. Calculate the activity of the client to be recommended on the patent trading platform; based on the corresponding relationship between the activity and the transaction intention score, obtain the first transaction intention score corresponding to the activity.
[0036] In this embodiment, the activity of the client to be recommended on the patent transaction platform is mainly based on the following two aspects: login duration and published purchase demand. By quantifying the landing time and the published purchase demand, the activity of the client to be recommended on the patent transaction platform is obtained.
[0037] Optionally, the login time is the average daily login time of the client to be recommended on the patent trading platform for the past three months. It should be understood that it can mainly reflect the frequent use of the patent transaction platform by the client to be recommended in the recent period. For example, the login time includes the average daily login time of the client to be recommended in the patent trading platform for K (1≤K≤6) months and/or the total login time of K (1≤K≤6) months, etc.
[0038] It should be understood that the longer the login time, the stronger the purchase demand of the customers to be recommended; the shorter the login time, the weaker the purchase demand. In this embodiment, the corresponding relationship between the login duration and the purchase demand level is constructed in advance to quantify the login duration. Among them, the corresponding relationship between the login duration and the purchase demand level is shown in Table 1 below:
[0039] Table 1
[0040] Login time t/min 0≤t≤10 10 20 30 40 Purchase demand level low Lower in Higher high
[0041] Different purchase demand levels correspond to different activity scores. For example, the "low" purchase demand level corresponds to an activity value of 10, the "lower" purchase demand level corresponds to an activity value of 20, and the "medium" purchase demand level corresponds to The activity value is 30, the activity value corresponding to the "higher" purchase demand level is 40, and the activity value corresponding to the "high" purchase demand level is 50. Thereby quantifying the duration of landing as activity.
[0042] It should be understood that the corresponding relationship between the login duration and the purchase demand level can be flexibly set based on actual needs, and is not limited to that shown in Table 1 above. The purchase demand level can be set to be more or less, and it is not limited to the above five levels of "low", "lower", "medium", "higher" and "high"; the range of login time corresponding to different levels can be Based on the shortest login time of each user of the patent trading platform t min , Average landing time t p , The longest login time t max Make flexible settings; for example, the "low" purchase demand level corresponds to the login duration range set to t min ≤t min+ t p )/2; "Medium" purchase demand level corresponds to the login duration range set to (t min+ t p )/2≤t p+ t max )/2; the "high" purchase demand level corresponds to the login duration range set to (t p+ t max )/2≤t max.
[0043] It is also possible to set the corresponding relationship between the login duration and the purchase demand level according to the ranking of the login duration distribution of each user of the patent transaction platform. For example, the login duration of each user of the patent trading platform is sorted in descending order, the top 20% users are determined, and the shortest login duration among the top 20% users is assumed to be t1, As the "high" purchase demand level corresponding to the login duration range, that is, t1≤t; then determine the top 40% of the users, the shortest login duration among the top 40% of the users, assume it is t2, as the The log-in duration range corresponding to the high purchase demand level, that is, t2≤t
[0044] The published purchase demand is a direct manifestation of the activeness of the customer to be recommended. When the current purchase demand of the customer to be recommended is larger, it indicates that the stronger the purchase demand of the customer to be recommended, the higher the activity; on the contrary, if the customer is to be recommended The smaller the purchase demand currently released by the customer, or even if the purchase demand is not published, the purchase demand of the customer to be recommended cannot be indicated, or the purchase demand of the customer to be recommended is weak or even no purchase demand.
[0045] Please see Table 2 below:
[0046] Table 2
[0047] Purchase demand released/article 0 1 2 ≥3 Activity value 0 30 40 50
[0048] Taking Table 1 and Table 2 above as examples, suppose that the average daily login time of the customer to be recommended in the past three months is 40 minutes, and the currently published purchase demand is one item. After quantitative processing, you can get the customer’s The activity level is 40+30=70. Then, based on the corresponding relationship between the activity and the transaction intention score, the first transaction intention score of the client to be recommended can be obtained. Here, the activity value ranges from 0 to 100. Assuming that the value of the transaction intention score ranges from 0 to 100, it can be converted at a ratio of 1:1 to convert the activity into a transaction intention score. That is, when the activity of the client to be recommended is 70, the corresponding first transaction intention score is 70 points. It should be understood that when the activity value range is different from the transaction intention score range, it can be flexibly set for conversion to accurately reflect the purchase intention degree of the customer to be recommended. The higher the transaction intention score, the higher the degree of purchase intention.
[0049] S102. Obtain patent information of the client to be recommended, and calculate a second transaction intention score based on the patent information; the patent information includes at least one of the current patent ownership, the patent purchase, and the growth rate of the patent purchase in the past three years.
[0050] The greater the current patent ownership of the client to be recommended, the greater the demand for the patent from the client to be recommended, the higher the emphasis on the patent, the stronger the purchase intention is usually; the amount of patent purchase is the strength of the client's intention to purchase the patent It is a direct manifestation of the fact that the more patents purchased, the stronger the purchase intention; the growth rate of the purchase of patents in the past three years can be used as the influencing factor of the patent purchase intention of the customers to be recommended, which can more accurately reflect the customer’s intentions purchase intention.
[0051] In this embodiment, the patent purchase volume is the average patent purchase volume in the past three years or the patent purchase volume in the previous year, which is not limited in other embodiments of the present invention.
[0052] Optionally, calculating the second transaction intention score based on patent information includes: first determining the intention score k1 corresponding to the current patent ownership of the client to be recommended based on the corresponding relationship between the patent ownership and the intention score; It is recommended that the growth rate of the patent purchase volume of the client in the past three years be revised, for example, the patent purchase volume is y1, and the growth rate of the patent purchase volume in the past three years is α, then the revised patent purchase volume is y=y1 *(1+α); Then for the revised patent purchase volume y, based on the corresponding relationship between the patent purchase volume and the intention score, determine the intention score k2 corresponding to the patent purchase volume of the client to be recommended; calculate k1+k2, The second transaction intention score corresponding to the patent information of the client to be recommended can be obtained.
[0053] S103. Obtain basic corporate information of the client to be recommended, and calculate a third transaction intention score based on the basic corporate information; the basic corporate information includes at least one of the company's scale, main business type, and operating status.
[0054] In this embodiment, the company scale includes at least one of the company's registered capital, paid-in capital, personnel scale, and number of insured persons; the larger the company scale, the stronger the company's financial strength.
[0055] The main business type refers to the main activities in the daily activities of the company to complete its business objectives. For example, for a VR company, its main business may usually include VR game software development, VR glasses products, etc.; for a communications company, its The main business may usually include base stations, servers, routers and other products; I will not repeat them here.
[0056] Due to the differences in the overall situation of different industries, there are differences in the transaction requirements and efficiency of patents in different industries. For example, the information transmission, computer services, and software industries have more patent requirements than manufacturing industries and can complete transactions faster; and manufacturing Compared with the financial industry, the industry has more demand for patents, so it can complete transactions faster. Based on this, this embodiment sets fixed intention scores corresponding to different industries. The higher the score, the easier it is to complete the transaction for patents in the industry, and the lower the score, the more difficult it is to complete the transaction. Specifically, it can be flexibly set after the overall analysis of the patent situation of each industry, and will not be repeated here.
[0057] Customers to be recommended generally have their own corresponding main business types. For example, the main business is blockchain products, and the corresponding industries are information transmission, computer services and software industries. It is assumed that the corresponding intention score is 10 points; assuming that the main business is For automobiles, the corresponding industry is manufacturing, assuming that the corresponding intention score is 8 points; assuming that the main business is pig breeding, the corresponding industry is animal husbandry, assuming that the corresponding intention score is 3 points, and so on.
[0058] Operating conditions include judicial administrative punishments and abnormal operating conditions. Judicial administrative punishments are, for example, the number of judicial administrative punishment decisions issued (including but not limited to the person subject to execution, consumption restriction, administrative punishment, equity pledge, movable property mortgage, etc.) , The number of abnormal business conditions such as being included in the list of business abnormalities.
[0059] It should be understood that when the client to be recommended has an abnormal operating condition, it will usually affect its patent purchase behavior. The fewer judicial administrative penalties and abnormal business situations of the client to be recommended, the higher the corresponding intention score. For example, there is no abnormal business situation, assuming the corresponding intention score is 10 points; there are 1 to 5 abnormal business operations Situation, assuming that the corresponding intention score is 8 points; there are 6-10 abnormal business conditions, assuming the corresponding intention score is 6 points; there are 11-20 abnormal business situations, assuming the corresponding intention score is 3 points; exist 21 and above abnormal business conditions, assuming the corresponding intention score is 0 points.
[0060] This embodiment fully combines basic information such as the company scale, main business type, and operating status of the client to be recommended to obtain the respective intention scores of the company scale, main business type, and operating status, and add them to obtain the third intention score. Value, which is conducive to more accurate analysis of the purchase intention of the customer to be recommended, thereby improving the accuracy of the recommendation.
[0061] It should be understood that the execution sequence of step S101 to step S103 can be flexibly adjusted, which is not limited in this embodiment.
[0062] S104: Calculate the weighted sum of the first transaction intention score, the second transaction intention score, and the third transaction intention score according to the respective weights of the activity, patent information, and basic enterprise information, as a recommendation index for the client to be recommended.
[0063] In this embodiment, the respective weights of activity, patent information, and basic enterprise information are 50%, 30%, and 20%, respectively. Assuming that the first transaction intention score obtained by the activity is Q1, the second transaction obtained by the patent information The intention score is Q2, and the third transaction intention score obtained from the basic information of the enterprise is Q3, then the recommended index Q=Q1*50%+Q2*30%+Q3*20%. It should be understood that the respective weights of activity, patent information, and basic enterprise information can be flexibly set according to actual conditions.
[0064] S105: Determine whether the recommendation index reaches the set threshold, if yes, go to step S106; if not, end.
[0065] In this embodiment, the threshold is set as the minimum recommendation index corresponding to all transaction customers. That is, the recommendation index of each customer with patent purchase records is calculated, and the smallest recommendation index value is selected as the set threshold described here.
[0066] S106. Regard the customer to be recommended as a customer with the intention of technology transfer to recommend to the corresponding seller to promote the transaction.
[0067] For the sellers recommended here, the technical fields of the patents to be traded on the shelves should meet the requirements of the technical fields in the release requirements of the customers of the technology transfer intention. It enables customers with high transaction intentions to directly match and connect with sellers who meet patent needs, so as to improve the efficiency of patent transactions and better meet the patent transaction needs of buyers and sellers.
[0068] According to the method for accurately recommending customers with technology transfer intentions provided by the present invention, the activity of the customers to be recommended on the patent transaction platform is calculated; based on the corresponding relationship between the activity and the transaction intention score, the first transaction intention score corresponding to the activity is obtained Value; Obtain patent information of the client to be recommended, and calculate the second transaction intention score based on the patent information; Patent information includes at least one of the current patent ownership, the number of patent purchases, and the growth rate of the number of patents purchased in the past three years; The company’s basic information of the client to be recommended is calculated based on the company’s basic information to obtain the third transaction intention score; the company’s basic information includes at least one of company size, main business type, and operating status; according to activity, patent information, and basic corporate information The weight corresponding to each information is calculated, and the weighted sum of the first transaction intention score, the second transaction intention score, and the third transaction intention score is calculated as the recommendation index of the client to be recommended; it is judged whether the recommendation index reaches the set threshold, if so, Customers to be recommended are regarded as customers with technology transfer intentions to recommend them to the corresponding sellers to promote transactions; the buyer and seller are bridged, and customers with strong purchase needs of technology transfer intentions are accurately recommended to the corresponding sellers, which promotes patent transactions and improves user experience.
