Abstract: In some embodiments, an intervention evaluation system estimates counterfactual metric for a focal online platform based on an assessment model built using performance data of the focal online platform and control online platforms. The intervention evaluation system accesses performance data of the focal online platform that has been subject to a temporary intervention and performance data of control online platforms that are not subject to the temporary intervention. The intervention evaluation system determines estimation weights for these control online platforms based on the performance data in a pre-intervention period. Based on the estimation weights, the intervention evaluation system computes a counterfactual metric indicating the performance of the focal online platform in a post-intervention period in the absence of the temporary intervention.

Abstract: Modifications to the DiD technique are disclosed which provide an estimate of the effectiveness of a site-wide action where no control group exists within the data subsequent to implementation of the site-wide action. In some examples, a method may include identifying a treatment group based on a modified treatment period, selecting a control group from a control period prior to the modified treatment period, and performing a modified difference-in-differences (DiD) estimation for a metric based on the modified treatment period, the treatment group, the control period, and the control group. The modified treatment period may encompass an intervention of a site-wide action, and include a pre-intervention time period and a post-intervention time period.

Abstract: In some embodiments, a computing system computes, with a state prediction model, probabilities of transitioning from a click state represented by interaction data to various predicted next states. The computing system computes an interface experience metric for the click with an experience valuation model. To do so, the computing system identifies base values for the click state and the predicted next states. The computing system computes value differentials for between the click state's base value and each predicted next state's base value. Value differentials indicate qualities of interface experience. The computing system determines the interface experience metric from a summation that includes the current click state's base value and the value differentials weighted with the predicted next states' probabilities.

Abstract: Systems, methods, and non-transitory computer-readable media are disclosed for easily, accurately, and efficiently determining a personalized market share of a user with a company versus that of its competitors using only focal company's own clickstream data. For instance, the disclosed systems can infer a mapping of purchases to product categories from clickstream data of a company and use the mappings to generate a dataset of observable conversions (with interconversion times) for one or more product categories. Then, the disclosed systems can utilize models for a category level interconversion time and for transition probabilities of a user to determine a personalized market share and an interconversion time for an individual user (between the company and competitors of the company).

Abstract: In some embodiments, an intervention evaluation system estimates counterfactual metric for a focal online platform based on an assessment model built using performance data of the focal online platform and control online platforms. The intervention evaluation system accesses performance data of the focal online platform that has been subject to a temporary intervention and performance data of control online platforms that are not subject to the temporary intervention. The intervention evaluation system determines estimation weights for these control online platforms based on the performance data in a pre-intervention period. Based on the estimation weights, the intervention evaluation system computes a counterfactual metric indicating the performance of the focal online platform in a post-intervention period in the absence of the temporary intervention.

Abstract: In some embodiments, a computing system computes, with a state prediction model, probabilities of transitioning from a click state represented by interaction data to various predicted next states. The computing system computes an interface experience metric for the click with an experience valuation model. To do so, the computing system identifies base values for the click state and the predicted next states. The computing system computes value differentials for between the click state's base value and each predicted next state's base value. Value differentials indicate qualities of interface experience. The computing system determines the interface experience metric from a summation that includes the current click state's base value and the value differentials weighted with the predicted next states' probabilities.

Abstract: This disclosure relates to methods, non-transitory computer readable media, and systems that, based on a sparse textual segment, can use machine learning models to generate document variants that are both conforming to digital content guidelines and uniquely tailored for distribution to client devices of specific audiences via specific delivery channels. To create such variants, in some embodiments, the methods, non-transitory computer readable media, and systems generate suggested modifications to a draft document that correspond to features of content-guideline-conforming documents. Additionally, or alternatively, in certain implementations, the disclosed methods, non-transitory computer readable media, and systems generate suggested modifications to a draft document that correspond to features of audience-channel-specific documents.

Abstract: Systems and techniques for identifying segments in categorical data include receiving multiple transaction ID (TID) lists with univariate values that satisfy a thresholding metric with each TID list representing an occurrence of a single attribute in a set of transactions. The TID lists are stored with the univariate values that satisfy the thresholding metric in a data structure. In a loop, candidate itemsets to form from combinations of TID lists are determined using only the combinations of TID lists that satisfy categorical constraints. In the loop, for the candidate itemsets that satisfy categorical constraints, both the thresholding metric and a similarity metric are applied to the candidate itemsets. Final itemsets are formed from only the candidate itemsets that satisfy both the thresholding metric and the similarity metric.

Abstract: This disclosure relates to utilizing a statistical model trained on character dimensions to determine a likelihood of a person purchasing a product. The method may include obtaining user-input data of a first person (e.g., textual-input data, survey-response data, offer information, or clickstream data associated with a first person). A character profile for the first person is derived using the user-input data and a psycholinguistic lexicon. A statistical model is generated based on the derived character profile of the first person. Second user-input data associated with a second person is obtained. The second user-input is applied to the statistical model to determine an output of the model (e.g., a statistical probability value that quantifies, for example, a predicted intention of the second person to purchase a particular product).

Abstract: Multimedia document summarization techniques are described. That is, given a document that includes text and a set of images, various implementations generate a summary by extracting relevant text segments in the document and relevant segments of images with constraints on the amount of text and number/size of images in the summary.

Abstract: In some embodiments, an intervention evaluation system estimates counterfactual metric for a focal online platform based on an assessment model built using performance data of the focal online platform and control online platforms. The intervention evaluation system accesses performance data of the focal online platform that has been subject to a temporary intervention and performance data of control online platforms that are not subject to the temporary intervention. The intervention evaluation system determines estimation weights for these control online platforms based on the performance data in a pre-intervention period. Based on the estimation weights, the intervention evaluation system computes a counterfactual metric indicating the performance of the focal online platform in a post-intervention period in the absence of the temporary intervention.

Abstract: There is described a method and system in an interactive computing environment modified with user experience values based on behavior logs. An experience valuation system determines an experience value and an estimated experience value. The experience value is based on a current state of interaction data from a user session, based on a history of past events, and an estimation function defined by parameters to model the user experience values. The estimated experience value is determined based on, in addition to the current state and the estimation function, next states associated with the current state, and a reward function. The parameters of the estimation function are updated based on a comparison of the expected experience value and the estimated experience value. For another aspect, the method and system may further include a state prediction system to determine probabilities of transitioning that may be applied to determine the estimated experience value.

Abstract: Systems and techniques for identifying segments in categorical data include receiving multiple transaction ID (TID) lists with univariate values that satisfy a thresholding metric with each TID list representing an occurrence of a single attribute in a set of transactions. The TID lists are stored with the univariate values that satisfy the thresholding metric in a data structure. In a loop, candidate itemsets to form from combinations of TID lists are determined using only the combinations of TID lists that satisfy categorical constraints. In the loop, for the candidate itemsets that satisfy categorical constraints, both the thresholding metric and a similarity metric are applied to the candidate itemsets. Final itemsets are formed from only the candidate itemsets that satisfy both the thresholding metric and the similarity metric.

Abstract: This disclosure relates to methods, non-transitory computer readable media, and systems that, based on a sparse textual segment, can use machine learning models to generate document variants that are both conforming to digital content guidelines and uniquely tailored for distribution to client devices of specific audiences via specific delivery channels. To create such variants, in some embodiments, the methods, non-transitory computer readable media, and systems generate suggested modifications to a draft document that correspond to features of content-guideline-conforming documents. Additionally, or alternatively, in certain implementations, the disclosed methods, non-transitory computer readable media, and systems generate suggested modifications to a draft document that correspond to features of audience-channel-specific documents.

Abstract: In some embodiments, a computing system computes, with a state prediction model, probabilities of transitioning from a click state represented by interaction data to various predicted next states. The computing system computes an interface experience metric for the click with an experience valuation model. To do so, the computing system identifies base values for the click state and the predicted next states. The computing system computes value differentials for between the click state's base value and each predicted next state's base value. Value differentials indicate qualities of interface experience. The computing system determines the interface experience metric from a summation that includes the current click state's base value and the value differentials weighted with the predicted next states' probabilities.

Abstract: Certain embodiments involve generating summarized versions of video content based on memorability of the video content. For example, a video summarization system accesses segments of an input video. The video summarization system identifies memorability scores for the respective segments. The video summarization system selects a subset of segments from the segments based on each computed memorability score in the subset having a threshold memorability score. The video summarization system generates visual summary content from the subset of the segments.

Abstract: Modifications to the DiD technique are disclosed which provide an estimate of the effectiveness of a site-wide action where no control group exists within the data subsequent to implementation of the site-wide action. In some examples, a method may include identifying a treatment group based on a modified treatment period, selecting a control group from a control period prior to the modified treatment period, and performing a modified difference-in-differences (DiD) estimation for a metric based on the modified treatment period, the treatment group, the control period, and the control group. The modified treatment period may encompass an intervention of a site-wide action, and include a pre-intervention time period and a post-intervention time period.

Abstract: Systems and methods disclosed select and display icons corresponding to topics of digital documents. In particular, in one or more embodiments, systems and methods identify similar clusters of documents from a plurality of documents and generate icons for each identified cluster of documents. For instance, systems and methods identify topical phrases from identified clusters and utilize the topical phrases to identify icons corresponding to the topical phrases. More specifically, one or more embodiments include systems and methods that generate generalized terms from topical phrases utilizing a concept taxonomy and then identify icons by comparing the generated generalized terms to tags corresponding to a plurality of icons.

Abstract: This disclosure relates to utilizing a statistical model trained on character dimensions to determine a likelihood of a person purchasing a product. The method may include obtaining user-input data of a first person (e.g., textual-input data, survey-response data, offer information, or clickstream data associated with a first person). A character profile for the first person is derived using the user-input data and a psycholinguistic lexicon. A statistical model is generated based on the derived character profile of the first person. Second user-input data associated with a second person is obtained. The second user-input is applied to the statistical model to determine an output of the model (e.g., a statistical probability value that quantifies, for example, a predicted intention of the second person to purchase a particular product).

Abstract: A digital medium environment is described to estimate a resulting causal impact of a particular item of digital marketing content on a digital marketing outcome. A first sequence is identified of a plurality of items of digital marketing content provided to a first set of users taken from a plurality of users. the first sequence includes the particular item. A second sequence is constructed of the plurality of items of the digital marketing content by removing the particular item from the first sequence. A second set is identified of users from the plurality of users provided with the second sequence. The resulting causal impact is estimated of the particular item of digital marketing content on the digital marketing outcome based on first and second causal impacts for subsets of the first and second sets.