Abstract: Disclosed is a blended gasoline composition having an AKI of 87. The formulation of the blended gasoline composition leads to a reduction in carbon dioxide emission. The blended gasoline composition contains a reduced concentration of olefins and non-amine aromatics.

Abstract: Disclosed is a blended gasoline composition having an AKI of 87. The formulation of the blended gasoline composition leads to a reduction in carbon dioxide emission. The blended gasoline composition contains a reduced concentration of olefins and non-amine aromatics.

Abstract: Example methods and systems for providing custom error pages for retail applications are disclosed. In general, the disclosed example systems provide error pages which are presented after a user attempts to access an inactive, unavailable, or nonexistent website address. In one aspect of the present disclosure, the system provides an error page which is customized, for example, based upon the website address that the user was attempting to access; based upon the user's history; based upon the system's most popular products and/or categories; and/or based upon some combination of those criteria.

Abstract: Disclosed are systems and methods for predicting demographics about authors of social media. A dataset relates features from social media messages to known demographic information about authors of the social media messages. The social media messages and known demographics may be retrieved from different social media services. Adjustable weights are associated with alternative demographic values for the features. The dataset is used to predict unknown demographics of users based on the adjustable weights associated with the alternative demographic values for extracted features from social media messages. A confidence level may be output for a predicted demographic that corresponds to a difference between a sum of weights associated with the predicted value and another demographic value. The predicted demographic value may be output when the confidence level exceeds a threshold. The weights may be updated when the confidence level is below the threshold even when the prediction is correct.

Abstract: Disclosed are systems and methods for predicting demographics about authors of social media. A dataset relates features from social media messages to known demographic information about authors of the social media messages. The social media messages and known demographics may be retrieved from different social media services. Adjustable weights are associated with alternative demographic values for the features. The dataset is used to predict unknown demographics of users based on the adjustable weights associated with the alternative demographic values for extracted features from social media messages. A confidence level may be output for a predicted demographic that corresponds to a difference between a sum of weights associated with the predicted value and another demographic value. The predicted demographic value may be output when the confidence level exceeds a threshold. The weights may be updated when the confidence level is below the threshold even when the prediction is correct.

Abstract: Example methods and systems for providing custom error pages for retail applications are disclosed. In general, the disclosed example systems provide error pages which are presented after a user attempts to access an inactive, unavailable, or nonexistent website address. In one aspect of the present disclosure, the system provides an error page which is customized, for example, based upon the website address that the user was attempting to access; based upon the user's history; based upon the system's most popular products and/or categories; and/or based upon some combination of those criteria.

Abstract: Disclosed herein is a memory system in which sought after keywords will identify their own location by setting a flag bit. The important thing is, keywords may be unstructured, hidden in a sea of words; finding one is equivalent to finding a needle in a haystack. To do so, keywords are given a flag bit that is initialized to zero, but conditionally flips true after running a brief algorithm. Subsequently, flag bits can be analyzed to provide the exact addresses of interesting data in a mass storage, separate from the memory in which text keywords are stored. A method of implementation, but not the only method, is to arrange cells in an integrated circuit to form a ‘word’ of memory, Each word may be structured to hold sought after keywords; it may also have optional ‘state’ data. A global bus enables the specification of any binary function on the keywords by selecting source bits in each word.