Abstract: In one embodiment, a cycling data platform may receive data corresponding to a cycling ride of a first user including one or more ride parameters. The cycling data platform may access a user profile of the first user comprising a cycling profile of the first user, where the cycling profile indicates one or more characteristics of cycling habits of the first user. The cycling data platform may update the cycling profile of the first user based on the one or more ride parameters. The cycling data platform may generate, using a recommendation engine, one or more recommendations for the first user based on the cycling profile of the first user.

Abstract: In one embodiment, a method comprises accessing a physician-referral-network, which comprises a number of nodes and a number of edges connecting the nodes, each of the edges representing a single degree of separation between the nodes. Each node represents a physician, and each edge represents a patient-referral between two physicians corresponding to the connected nodes. One or more references indicating a patient-referral from a first physician to a second physician is received. The physician-referral-network is updated based on the received reference. One or more performance-scores from the second physicians may be received, each performance-score corresponding to a patient-referral. A referral-score is calculated for the first physician based on the performance-scores. The method further comprises determining if a referral-score for a first physician is below a threshold referral-score.

Abstract: Systems, methods, and software described herein provide enhancements for managing and comparing data from multiple information source. In one implementation, a method of operating a summary service includes identifying a user request for a summary of an event, generating the summary using scores for each data source for the event, and providing the summary to the requesting user. In some implementations, the method may further include obtaining, from the data sources, data objects in a plurality of formats corresponding to the event, wherein the data objects each comprise data items for the event, and maintaining a data structure for the event, wherein the data structure relates data items of similar data type from the plurality of data sources.

Abstract: In one embodiment, a method comprises identifying one or more health-care-providers associated with one or more medical-specializations, wherein each medical-specialization is associated with a specialty class and at least one geographic area. A number of patient-diagnosis codes is received, each patient-diagnosis code corresponding to one of the health-care-providers and representing a patient-visit to the health-care-provider. The patient-diagnosis code includes at least one diagnosis within a condition-group for the patient visit, and a severity-factor for the diagnosis. The method further comprises determining all patient-diagnosis codes corresponding to the health-care-providers. A patient-volume is calculated for each health-care-provider, based on the patient-diagnosis codes and associated severity-factors. A variety-score is calculated for each health-care-provider based on the patient-diagnosis codes.

Abstract: In one embodiment, a method comprises accessing a physician-referral-network, which comprises a number of nodes and a number of edges connecting the nodes, each of the edges representing a single degree of separation between the nodes. Each node represents a physician, and each edge represents a patient-referral between two physicians corresponding to the connected nodes. One or more references indicating a patient-referral from a first physician to a second physician is received. The physician-referral-network is updated based on the received reference. One or more performance-scores from the second physicians may be received, each performance-score corresponding to a patient-referral. A referral-score is calculated for the first physician based on the performance-scores. The method further comprises determining if a referral-score for a first physician is below a threshold referral-score.

Abstract: In one embodiment, a method comprises receiving a set of parameters for selecting a physician from an administrator of a matching-engine system. The matching-engine system stores the parameters in a data store of the matching-engine system. The matching-engine system receives from a user a search query with a location of the user and a user-specified symptom or treatment. The matching-engine system determines a base-concept of a medical diagnosis or medical procedure based on the search query. The matching-engine system may identify a set of physicians to be recommended to the user based on the parameters, and a performance-score and experience-score associated with the base-concept for each physician. The matching-engine system may send a search-results page to the user listing the recommended physicians.

Abstract: In one embodiment, a method comprises accessing a physician-referral-network, which comprises a number of nodes and a number of edges connecting the nodes, each of the edges representing a single degree of separation between the nodes. Each node represents a physician, and each edge represents a patient-referral between two physicians corresponding to the connected nodes. One or more references indicating a patient-referral from a first physician to a second physician is received. The physician-referral-network is updated based on the received reference. One or more performance-scores from the second physicians may be received, each performance-score corresponding to a patient-referral. A referral-score is calculated for the first physician based on the performance-scores. The method further comprises determining if a referral-score for a first physician is below a threshold referral-score.

Abstract: In one embodiment, a method comprises receiving an indication of one or more care-episodes from a client device. Each care-episode may include an identifier for one or more physicians, a base-concept corresponding to a medical condition or medical treatment, one or more sub-concepts associated with the base-concept, and an indication of the resources used during the care-episode. The method comprises calculates one or more cost-factors for a base-concept for a particular physician based on the care-episodes associated with the base-concept. The method further comprises determining a specialty and geographic region of the particular physician. A peer group of physicians in the same specialty and geographic region is selected. A performance score is calculated based on the cost-factor for the physician compared to the cost-factors for the peer group of physicians.

Abstract: In one embodiment, a method comprises receiving a set of parameters for selecting a physician from an administrator of a matching-engine system. The matching-engine system stores the parameters in a data store of the matching-engine system. The matching-engine system receives from a user a search query with a location of the user and a user-specified symptom or treatment. The matching-engine system determines a base-concept of a medical diagnosis or medical procedure based on the search query. The matching-engine system may identify a set of physicians to be recommended to the user based on the parameters, and a performance-score and experience-score associated with the base-concept for each physician. The matching-engine system may send a search-results page to the user listing the recommended physicians.

Abstract: In one embodiment, a method comprises identifying one or more health-care-providers associated with one or more medical-specializations, wherein each medical-specialization is associated with a specialty class and at least one geographic area. A number of patient-diagnosis codes is received, each patient-diagnosis code corresponding to one of the health-care-providers and representing a patient-visit to the health-care-provider. The patient-diagnosis code includes at least one diagnosis within a condition-group for the patient visit, and a severity-factor for the diagnosis. The method further comprises determining all patient-diagnosis codes corresponding to the health-care-providers. A patient-volume is calculated for each health-care-provider, based on the patient-diagnosis codes and associated severity-factors. A variety-score is calculated for each health-care-provider based on the patient-diagnosis codes.