http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-106897911-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_a431cf4c64da0f89eccaa9aa30b43b7f |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06Q30-0631 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-23 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06Q30-06 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
| filingDate | 2017-01-10^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_41f6c2ce3e7deda184ae39e2cea7624f http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_eb319ae97258471f550c838333d62e4f |
| publicationDate | 2017-06-27^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-106897911-A |
| titleOfInvention | An Adaptive Personalized Recommendation Method Based on Users and Items |
| abstract | The invention discloses an adaptive personalized recommendation method based on users and items, which is divided into two stages of training and personalized recommendation. In the training phase, the platform first collects data such as user personal information, user behavior characteristics, and user evaluation of items. According to the user data, similar users are clustered, and the average difference matrix of the user's evaluation of the item is calculated. Based on this, a prediction model based on user clustering is established, and the evaluation prediction error of the model for all items is calculated; Calculate the average difference of the user's evaluation of the item, establish a prediction model, and form an adaptive prediction model based on users and items. In the personalized recommendation stage, the cluster to which the user belongs is first judged, and the user's evaluation of the item is estimated by using an adaptive prediction model that integrates the user and the item, and the item with a high predicted evaluation is recommended to the user. Compared with the traditional personalized recommendation method, the present invention has self-adaptive ability and higher accuracy. |
| isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111256303-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-107679945-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114648391-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111256303-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110321490-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110069663-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110069663-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/JP-7245904-B2 http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-108334592-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111611496-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-108334592-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-108595598-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114648391-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-109903103-A |
| priorityDate | 2017-01-10^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Showing number of triples: 1 to 32 of 32.