Abstract: Techniques for predicting a user response to content are described. According to various embodiments, a configuration file is accessed, where the configuration file includes a user-specification of raw data accessible via external data sources and raw data encoding rules. In some embodiments, the raw data includes raw member data associated with a particular member and raw content data associated with a particular content item. Thereafter, source modules encode the raw data from the external data sources into feature vectors, based on the raw data encoding rules. An assembler module assembles one or more of the feature vectors into an assembled feature vector, based on user-specified assembly rules included in the configuration file. A prediction module performs a prediction modeling process based on the assembled feature vector and a prediction model, to predict a likelihood of the particular member performing a particular user action on the particular content item.

Abstract: An ad player presents ads in association with a video player by evaluating an associated ad script. The ad player transforms data included in the ad script into operational instructions. Hence, the ad player flexibly and dynamically configures itself and presents ads in accordance with the contents of the ad script, enabling a publisher to modify advertising aspects simply by modifying the ad script. The ad script can comprise a script in a tag-based markup language that is readable by the ad player. For example, the ad script can include one or more tags, each tag including one or more attributes that are each set to a value. The ad player determines the values of the attributes and presents ads in accordance with associated ad characteristics or behaviors.

Abstract: An ad player presents ads in association with a video player by evaluating an associated ad script. The ad player transforms data included in the ad script into operational instructions. Hence, the ad player flexibly and dynamically configures itself and presents ads in accordance with the contents of the ad script, enabling a publisher to modify advertising aspects simply by modifying the ad script. The ad script can comprise a script in a tag-based markup language that is readable by the ad player. For example, the ad script can include one or more tags, each tag including one or more attributes that are each set to a value. The ad player determines the values of the attributes and presents ads in accordance with associated ad characteristics or behaviors.

Abstract: Techniques for predicting a user response to content are described. According to various embodiments, a configuration file is accessed, where the configuration file includes a user-specification of raw data accessible via external data sources and raw data encoding rules. In some embodiments, the raw data includes raw member data associated with a particular member and raw content data associated with a particular content item. Thereafter, source modules encode the raw data from the external data sources into feature vectors, based on the raw data encoding rules. An assembler module assembles one or more of the feature vectors into an assembled feature vector, based on user-specified assembly rules included in the configuration file. A prediction module performs a prediction modeling process based on the assembled feature vector and a prediction model, to predict a likelihood of the particular member performing a particular user action on the particular content item.

Abstract: Techniques for predicting a user response to content are provided. In example embodiments, one or more feature vectors are assembled into an assembled feature vector. A particular one of the feature vectors not being available is determined. In response to determining that the particular one of the feature vectors is not available, the particular feature vector is ignored based on an importance value associated with the particular feature vector. A substitute value associated with the particular feature vector is inserted into a portion of the assembled feature vector associated with the particular feature vector. A prediction modeling process based on the assembled feature vector and a prediction model is performed to predict a likelihood of a particular member performing a particular user action on a particular content item.

Abstract: Techniques for predicting a user response to content are described. According to various embodiments, a configuration file is accessed, where the configuration file includes a user-specification of raw data accessible via external data sources and raw data encoding rules. In some embodiments, the raw data includes raw member data associated with a particular member and raw content data associated with a particular content item. Thereafter, source modules encode the raw data from the external data sources into feature vectors, based on the raw data encoding rules. An assembler module assembles one or more of the feature vectors into an assembled feature vector, based on user-specified assembly rules included in the configuration file. A prediction module performs a prediction modeling process based on the assembled feature vector and a prediction model, to predict a likelihood of the particular member performing a particular user action on the particular content item.

Abstract: Techniques for predicting a user response to content are described. According to various embodiments, a configuration file is accessed, where the configuration file includes a user-specification of raw data accessible via external data sources and raw data encoding rules. In some embodiments, the raw data includes raw member data associated with a particular member and raw content data associated with a particular content item. Thereafter, source modules encode the raw data from the external data sources into feature vectors, based on the raw data encoding rules. An assembler module assembles one or more of the feature vectors into an assembled feature vector, based on user-specified assembly rules included in the configuration file. A prediction module performs a prediction modeling process based on the assembled feature vector and a prediction model, to predict a likelihood of the particular member performing a particular user action on the particular content item.

Abstract: Techniques for predicting a user response to content are described. According to various embodiments, a configuration file is accessed, where the configuration file includes a user-specification of raw data accessible via external data sources and raw data encoding rules. In some embodiments, the raw data includes raw member data associated with a particular member and raw content data associated with a particular content item. Thereafter, source modules encode the raw data from the external data sources into feature vectors, based on the raw data encoding rules. An assembler module assembles one or more of the feature vectors into an assembled feature vector, based on user-specified assembly rules included in the configuration file. A prediction module performs a prediction modeling process based on the assembled feature vector and a prediction model, to predict a likelihood of the particular member performing a particular user action on the particular content item.

Abstract: A method of matching advertisements to users is disclosed. A plurality of attributes of a population of users is identified. A selection is received of an attribute of the plurality of attributes to which a target value is to be broadly matched. A correspondence of an advertisement to a user is determined based on a broad matching of the target value to the attribute. The advertisement is matched to the user based at least in part on the determining of the correspondence.