Abstract: The present disclosure is directed toward systems and methods for identifying contributing factors associated with a multi-variable metric anomaly. One or more embodiments described herein identify one or more contributing factors that led to an anomaly in a multi-variable metric by calculating linearizing weights such that the total deviation in the multi-variable metric can be written as a weighted sum of deviations for dimension elements associated with the multi-variable metric.

Abstract: Computer-implemented methods and systems are disclosed for making a recommending providing a post on a social media forum. One exemplary embodiment involves utilizing machine-learning techniques to produce a model capable of determining optimal post recommendations from various posting factors. The model may be produced from historical post information regarding various posts made by, for instance, marketers on a social media forum and corresponding community interest responses to the posts made by the community of users associated with the social media forum. The model may be provided to a recommendation engine.

Abstract: Determined seed groups herein improve content dissemination across a communication network connecting a plurality of users. Probabilities of each user in the plurality influencing remaining users in the plurality to observe the content are identified to select a first influential user from the plurality. The seed group size is established and a user of the plurality with a probability proximate to the probability of the first influential user is identified. Based on the seed group size, the probabilities of the remaining users in the plurality are unified with the probability of the first influential user to determine new probabilities of the remaining users, and another user of the plurality with a probability proximate to the probability of the first influential user is identified. The method then provides for selecting the users identified as having probabilities proximate to the probability of the first influential user to establish the seed group.

Abstract: Techniques described herein relate to calculating the effectiveness or marketing “lift” of online social media promotions (e.g., Tweets® made on Twitter® or postings made on Facebook®), based on the impact that any such promotion is measured to have, after the promotion is made. Key performance indicators (KPI) for online social media marketing efforts may be established or updated based on such calculations. The techniques disclosed herein may also provide a direct way of measuring impact of an online social media promotion on brand awareness among an online social media audience. This may be accomplished while taking into account any effects of other promotions that have been made or are being made contemporaneously on the online social media platform or other non-social media marketing efforts.

Abstract: Techniques for multi-channel marketing campaigns are described herein. The techniques enable marketers to determine sequences of chronologically ordered communication channels by which to perform a multi-channel marketing campaign. In some cases, the techniques determine a sequence likely to have a positive result based on historic marketing sequence data and a desired category of the marketer's campaign. The techniques may also determine some number of trial sequences for a trial marketing campaign and then determine a best sequence for a full-scale marketing campaign based on the success of the trial sequences during the trial marketing campaign.

Abstract: Techniques to personalize a sequence of marketing actions and/or marketing channels used to contact individuals are described herein. Marketing data may be analyzed to select a sequence of marketing actions to employ for targeted marketing to an individual user. The analysis involves a comparison of sequence data obtained from collected marketing data that describes sequencing for the marketing offers provided to consumers to one or more potential sequences for the individual user. The potential sequences may be ranked based on similarities in characteristics of consumers associated with sequences that achieved a designated objective and the individual user's marketing sequence. Characteristics used for the ranking may further include demographic details and behavioral information regarding the consumers and individual user. One or more top ranking sequences are identified and employed to determine one or more marketing actions to perform next to provide targeted marketing offers to the individual user.

Abstract: Techniques for preference mapping for automated attribute selection in campaign design are described. In one or more implementations, consumer preference data associated with a plurality of products including a client product is analyzed by one or more computing devices to determine user sentiments associated with attributes that correspond to respective products. In addition, scores are assigned to the attributes based on the user sentiments associated with the attributes. Then, a preference mapping is performed using the assigned scores to generate a displayable representation of a comparison between at least two of the plurality of products based on the consumer preference data and a relative proximity of each attribute to corresponding products with respect to associated user sentiment. Subsequently, the displayable representation is communicated such that the displayable representation is identifiable regarding which attributes of the client product to highlight in a marketing campaign.

Abstract: Techniques for automatic identification of sources of web metric changes are described. In one or more implementations, changes in a web metric that indicate a measurable attribute associated with a website are determined, and the web metric is analyzed to identify sources that contributed to the changes in the web metric. In implementations, data is queried to obtain actual values for dimension elements along one or more dimensions of the web metric. In addition, expected values for the dimension elements are estimated along the dimensions of the web metric based on historical data. Then, deviations between the actual values and the expected values are calculated by using comparable statistics. Subsequently, the comparable statistics can be analyzed to identify corresponding dimension elements as the sources that contributed to the changes in the web metric.

Abstract: Techniques are disclosed for evaluating the incremental effect of a marketing channel that forms part of a multichannel marketing campaign. In one implementation data characterizing observed marketing interactions and outcomes is collected. A conversion probability is estimated as a function of the observed interactions using logistic regression techniques, wherein converting and non-converting consumers comprise the two classes upon which the regression is based. As a result, marketing interactions that are relatively more commonplace amongst converting consumers (as compared to non-converting consumers) receive greater attribution for observed conversions. The estimated conversion probability is then used to predict an incremental quantity of conversions that can be attributed to a kth marketing channel based on the average treatment effect.

Abstract: This document describes techniques for predicting reactions to short-text posts. In one or more implementations, a prediction model for short-text posts is generated from previous posts to a social network and responses to the posts by the social network community. Subsequently, the prediction model can be used to predict the social network community's reaction to a proposed post prior to the proposed post being posted to the social network.

Abstract: Techniques for providing information about large data sets may be provided. For example, a summary of the data sets and of patterns between the data sets may be presented. Traffic associated with a network-based resource that includes a number of documents may be an example of large data sets. The traffic may be analyzed and traffic patterns may be determined. A structure may be generated based on the traffic patterns and may use nodes to represent the documents. Further, a visualization of the structure may be presented. The visualization may include recursive clusters of the nodes, where the clusters may be labeled based on the respective clustered nodes.

Abstract: Computer-implemented methods and systems are disclosed for making a recommending providing a post on a social media forum. One exemplary embodiment involves utilizing machine-learning techniques to produce a model capable of determining optimal post recommendations from various posting factors. The model may be produced from historical post information regarding various posts made by, for instance, marketers on a social media forum and corresponding community interest responses to the posts made by the community of users associated with the social media forum. The model may be provided to a recommendation engine.

Abstract: This document describes techniques for predicting reactions to short-text posts. In one or more implementations, a prediction model for short-text posts is generated from previous posts to a social network and responses to the posts by the social network community. Subsequently, the prediction model can be used to predict the social network community's reaction to a proposed post prior to the proposed post being posted to the social network.

Abstract: Determined seed groups herein improve content dissemination across a communication network connecting a plurality of users. Probabilities of each user in the plurality influencing remaining users in the plurality to observe the content are identified to select a first influential user from the plurality. The seed group size is established and a user of the plurality with a probability proximate to the probability of the first influential user is identified. Based on the seed group size, the probabilities of the remaining users in the plurality are unified with the probability of the first influential user to determine new probabilities of the remaining users, and another user of the plurality with a probability proximate to the probability of the first influential user is identified. The method then provides for selecting the users identified as having probabilities proximate to the probability of the first influential user to establish the seed group.

Abstract: Embodiments of the present invention disclose a method and system for determining customer conversion propensity. According to one embodiment, a source registrant group having a plurality of registered customers is identified from a database. A probability for a conversion event is then computed for each of the plurality of customers within the source registrant group is then computed. The customers within the source registrant are than categorized based on the computed probability value and a timing for the conversion event.