Abstract: Machine-learning-aided automatic taxonomy for web data. In an embodiment, a training dataset of annotated features is used to train a model to predict a class in a taxonomy of web-based activities. The features may be derived from a uniform resource locator (URL) of an online resource and associated metadata. During operation, the features may be extracted from the URL and metadata of each activity record in web data. The trained model may be applied to the extracted features for each activity record to predict a class within the taxonomy. The predicted taxonomic class may be stored in association with the URL that was extracted from the activity record to produce a taxonomized URL.

Abstract: Aggregation of noisy datasets into a master firmographic database. In an embodiment, firmographic records are received from a plurality of sources, and normalized into a common schema. One or more firmographic records may be cleaned by replacing a value of one or more fields in those firmographic record(s) with a value of those field(s) in another firmographic record. The firmographic records may then be clustered, and each of the clusters may be collapsed into a single conflated firmographic record based on a voting process. A master identifier may be generated for each conflated firmographic record, and the conflated firmographic records may be merged into a master firmographic database that is indexed by master identifiers.

Abstract: Currently, significant technological obstacles prevent the automation of campaign configuration. Embodiments overcome these obstacles using an artificial intelligence (AI) model (e.g., comprising a k-nearest neighbor algorithm) that determines historical segments that match a target segment, based on one or more segment attributes. One or more thresholding criteria of a performance filter are applied to campaign configurations, associated with these matching historical segments, to determine one or more campaign configurations to be recommended. If one of these recommended campaign configurations is accepted, campaign settings for a new campaign may be auto-populated according to the accepted campaign configuration, such that a new campaign, with expected high performance, can be easily configured.

Abstract: Machine-learning-aided automatic taxonomy for web data. In an embodiment, a training dataset of annotated features is used to train a model to predict a class in a taxonomy of web-based activities. The features may be derived from a uniform resource locator (URL) of an online resource and associated metadata. During operation, the features may be extracted from the URL and metadata of each activity record in web data. The trained model may be applied to the extracted features for each activity record to predict a class within the taxonomy. The predicted taxonomic class may be stored in association with the URL that was extracted from the activity record to produce a taxonomized URL.

Abstract: Computer implemented sales prediction system collects data relating to events of visitors showing an interest in a client company from plural data sources, an organization module which organizes collected data into different event types and separates collected event counts in each event type between non-recent and recent events occurring within a predetermined time period, a first processing module which periodically calculates weighting for each event type based on recent and non-recent events for the event type compared to totals for other selected event types, a second processing module which periodically calculates sales prediction scores for each visitor and companies with which visitors are associated based on accumulated event data and weighting, and a reporting and data extract module which is configured to detect variation in sales prediction scores over time to identify spikes which can predict upcoming sales and to provide predicted sales information and leads to the client company.

Abstract: Aggregation of noisy datasets into a master firmographic database. In an embodiment, firmographic records are received from a plurality of sources, and normalized into a common schema. One or more firmographic records may be cleaned by replacing a value of one or more fields in those firmographic record(s) with a value of those field(s) in another firmographic record. The firmographic records may then be clustered, and each of the clusters may be collapsed into a single conflated firmographic record based on a voting process. A master identifier may be generated for each conflated firmographic record, and the conflated firmographic records may be merged into a master firmographic database that is indexed by master identifiers.

Abstract: Mapping anonymous Internet entities to known accounts. In an embodiment, events, representing online activity and comprising IP addresses, are received from a plurality of sources. Subsets of the events are aggregated into mappings that associate the IP address, shared by the subset, with an account. Each mapping is associated with statistics regarding the events. A confidence value is calculated for each mapping based on the statistics, and a final subset of the mappings is selected based on the confidence values. Subsequently, when a request with an IP address is received, the final subset of mappings is searched for the requested IP address, and an indication of the account associated with the requested IP address is returned in response to the request.

Abstract: Mapping anonymous Internet entities to known accounts. In an embodiment, events, representing online activity and comprising IP addresses, are received from a plurality of sources. Subsets of the events are aggregated into mappings that associate the IP address, shared by the subset, with an account. Each mapping is associated with statistics regarding the events. A confidence value is calculated for each mapping based on the statistics, and a final subset of the mappings is selected based on the confidence values. Subsequently, when a request with an IP address is received, the final subset of mappings is searched for the requested IP address, and an indication of the account associated with the requested IP address is returned in response to the request.

Abstract: Computer implemented sales prediction system collects data relating to events of visitors showing an interest in a client company from plural data sources, an organization module which organizes collected data into different event types and separates collected event counts in each event type between non-recent and recent events occurring within a predetermined time period, a first processing module which periodically calculates weighting for each event type based on recent and non-recent events for the event type compared to totals for other selected event types, a second processing module which periodically calculates sales prediction scores for each visitor and companies with which visitors are associated based on accumulated event data and weighting, and a reporting and data extract module which is configured to detect variation in sales prediction scores over time to identify spikes which can predict upcoming sales and to provide predicted sales information and leads to the client company.

Abstract: Mapping anonymous Internet entities to known accounts. In an embodiment, events, representing online activity and comprising IP addresses, are received from a plurality of sources. Subsets of the events are aggregated into mappings that associate the IP address, shared by the subset, with an account. Each mapping is associated with statistics regarding the events. A confidence value is calculated for each mapping based on the statistics, and a final subset of the mappings is selected based on the confidence values. Subsequently, when a request with an IP address is received, the final subset of mappings is searched for the requested IP address, and an indication of the account associated with the requested IP address is returned in response to the request.

Abstract: Mapping anonymous Internet entities to known accounts. In an embodiment, events, representing online activity and comprising IP addresses, are received from a plurality of sources. Subsets of the events are aggregated into mappings that associate the IP address, shared by the subset, with an account. Each mapping is associated with statistics regarding the events. A confidence value is calculated for each mapping based on the statistics, and a final subset of the mappings is selected based on the confidence values. Subsequently, when a request with an IP address is received, the final subset of mappings is searched for the requested IP address, and an indication of the account associated with the requested IP address is returned in response to the request.

Abstract: Computer implemented sales prediction system collects data relating to events of visitors showing an interest in a client company from plural data sources, an organization module which organizes collected data into different event types and separates collected event counts in each event type between non-recent and recent events occurring within a predetermined time period, a first processing module which periodically calculates weighting for each event type based on recent and non-recent events for the event type compared to totals for other selected event types, a second processing module which periodically calculates sales prediction scores for each visitor and companies with which visitors are associated based on accumulated event data and weighting, and a reporting and data extract module which is configured to detect variation in sales prediction scores over time to identify spikes which can predict upcoming sales and to provide predicted sales information and leads to the client company.

Abstract: A computer implemented sales prediction system collects data relating to events of visitors showing an interest in a client company from plural data sources, an organization module which organizes collected data into different event types and separates the collected event counts in each event type between non-recent events and recent events occurring within a predetermined time period, a first processing module which periodically calculates weighting for each event type based on recent events and non-recent events for the event type compared to totals for other selected event types, a second processing module which periodically calculates sales prediction scores for each visitor and companies with which visitors are associated based on the accumulated event data and weighting, and a reporting and data extract module which is configured to detect variation in sales prediction scores over time to identify spikes which can predict upcoming sales and to provide predicted sales information and leads to the client