Intellectual property recommending method and system
View Patent ↗An intellectual property (IP) recommending method and an IP recommending system are provided. In the method, a plurality of IP portfolios respectively designated for a plurality of product designs are retrieved and usage data of a plurality of IPs included in each of the plurality of IP portfolios are extracted. A machine learning (ML) model is trained by using a portion of the retrieved IP portfolios and the extracted usage data. In response to receiving at least one criterion for a desired product design from a user, a plurality of IPs adapted for the desired product design are predicted based on the ML model and recommended for the user.
1. An intellectual property (IP) recommending method for an electronic apparatus having a processor, the method comprising:
retrieving, by the processor, a plurality of IP portfolios respectively designated for a plurality of product designs comprising a tap-out in the semiconductor industry and product data related to a product of each of the product designs;
extracting, by the processor, usage data of a plurality of IPs included in each of the plurality of IP portfolios, wherein the usage data comprises information of each of the IPs in production and the product the IP is applied for, and the product data comprises one or a combination of market segment information;
training, by the processor, a machine learning (ML) model to learn relationships among market segment of the product and usage of the IP by using a portion of the retrieved plurality of IP portfolios, the retrieved product data and the extracted usage data;
in response to receiving at least one criterion for a desired product design from a user, predicting, by the processor, a plurality of IPs for the desired product design based on the ML model; and
recommending, by the processor, the predicted plurality of IPs for the user by displaying the predicted plurality of IPs for selection.
2. The method of claim 1 , wherein the step of retrieving the plurality of IP portfolios comprises:
labeling each of the plurality of IPs included in each of the plurality of IP portfolios with at least one criterion, wherein the at least one criterion comprises one or a combination of a provider, a name, a type and a version of the IP, and a dimension, a process, a protocol, a performance, and a power consumption required for a product of the IP.
3. The method of claim 1 , wherein the step of extracting usage data of a plurality of IPs comprises:
labeling the usage data of each of the plurality of IPs with at least one user adopting the IP and at least one product the IP is applied for.
4. The method of claim 1 , further comprising:
displaying a user interface including at least one entry for receiving an input of the at least one criterion for the desired product design; and
displaying the predicted plurality of IPs recommended for the user on the user interface.
5. The method of claim 1 , wherein the step of displaying the predicted plurality of IPs comprises:
displaying the predicted plurality of IPs as a list including fields of one or a combination of a name of each of the IPs, and a type of a product of each of the IPs.
6. The method of claim 1 , wherein the ML model comprises one or a combination of a rule-based algorithm, a segment-based algorithm, an item-based algorithm, or a segment and item based algorithm.
7. The method of claim 1 , further comprising:
evaluating a performance of the trained ML model by using a remaining portion of the retrieved plurality of IP portfolios and the extracted usage data.
8. The method of claim 1 , further comprising:
receiving the IPs selected from the recommended IPs by the user; and
training the ML model by using the received at least one criterion for the desired product design and the received IPs.
9. The method of claim 1 , further comprising:
sorting the predicted plurality of IPs according to a usage probability of each of the predicted plurality of IPs.
10. An intellectual property (IP) recommending system, comprising:
a data retrieving device, configured to connect and retrieve at least one IP database;
a storage device, configured to store data retrieved by the data retrieving device;
an input device, configured to receive an input of a user;
a display; and
a processor, coupled to the data retrieving device, the storage device, the input device and the display, and configured to:
retrieve a plurality of IP portfolios respectively designated for a plurality of product designs comprising a tap-out in the semiconductor industry and product data related to a product of each of the product designs from the at least one IP database by using the data retrieving device;
extract usage data of a plurality of IPs included in each of the plurality of IP portfolios, wherein the usage data comprises information of each of the IPs in production and the product the IP is applied for, and the product data comprises one or a combination of market segment information;
train a machine learning (ML) model to learn relationships among market segment of the product and usage of the IP by using a portion of the retrieved plurality of IP portfolios, the retrieved product data and the extracted usage data and store a plurality of parameters of the trained ML model in the storage device;
in response to the input device receiving at least one criterion for a desired product design from the user, predict a plurality of IPs for the desired product design based on the ML model; and
recommend the predicted plurality of IPs for the user by displaying the predicted plurality of IPs for selection on the display.
11. The IP recommending system of claim 10 , wherein the processor further labels each of the plurality of IPs included in each of the plurality of IP portfolios with at least one criterion, wherein the at least one criterion comprises one or a combination of a provider, a name, a type and a version of the IP, and a dimension, a process, a protocol, a performance, and a power consumption required for a product of the IP.
12. The IP recommending system of claim 10 , wherein the processor further labels the usage data of each of the plurality of IPs with at least one user adopting the IP and at least one product the IP is applied for.
13. The IP recommending system of claim 10 , wherein the processor further displays a user interface including at least one entry on the display for receiving an input of the at least one criterion for the desired product design, and displays the predicted plurality of IPs recommended for the user on the user interface on the display.
14. The IP recommending system of claim 10 , wherein the processor further displays the predicted plurality of IPs as a list including fields of one or a combination of a name and a provider of each of the IPs, and a type of a product of each of the IPs.
15. The IP recommending system of claim 10 , wherein the ML model comprises a rule-based algorithm, a segment-based algorithm, an item-based algorithm, or a segment and item based algorithm.
16. The IP recommending system of claim 10 , wherein the processor further evaluates a performance of the trained ML model by using a remaining portion of the retrieved plurality of IP portfolios and the extracted usage data.
17. The IP recommending system of claim 10 , wherein the processor further receives the IPs selected from the recommended IPs by the user by the input device, and trains the ML model by using the received at least one criterion for the desired product design and the received IPs.
18. A non-transitory computer readable medium, storing programs to be loaded into an electronic device having a processor, to perform steps of:
retrieving, by the processor, a plurality of IP portfolios respectively designated for a plurality of product designs comprising a tap-out in the semiconductor industry and product data related to a product of each of the product designs;
extracting, by the processor, usage data of a plurality of IPs included in each of the plurality of IP portfolios, wherein the usage data comprises information of each of the IPs in production and the product the IP is applied for, and the product data comprises one or a combination of market segment information;
training, by the processor, a machine learning (ML) model to learn relationships among market segment of the product and usage of the IP by using a portion of the retrieved plurality of IP portfolios, the retrieved product data and the extracted usage data;
in response to receiving at least one criterion for a desired product design from a user, predicting, by the processor, a plurality of IPs for the desired product design based on the ML model; and
recommending, by the processor, the predicted plurality of IPs for the user by displaying the predicted plurality of IPs for selection.
19. The non-transitory computer readable medium of claim 18 , further comprising:
displaying a user interface including at least one entry for receiving an input of the at least one criterion for the desired product design; and
displaying the predicted plurality of IPs recommended for the user on the user interface.
20. The non-transitory computer readable medium of claim 18 , wherein the step of displaying the predicted plurality of IPs comprises:
displaying the predicted plurality of IPs as a list including fields of one or a combination of a name of each of the IPs, and a type of a product of each of the IPs.