Abstract: A method, system and computer program product for database search enhancement and interactive user interface therefor. Database records are ranked by a match score, calculated using a plurality of criteria for determining a match level between values of a record and a search query, and using a plurality of priority parameters for aggregating the match level determined. A plurality of top-ranking records is selected, and a diversity measure is calculated therefor, using at least one class label assigned to records therein. If a sufficiency condition is not met by the diversity measure, at least one reference set of records sharing a class label in common is extracted from the plurality of top-ranking records and analyzed for determining at least one modification to the search query in improvement of the diversity measure, the scoring computational operator is accordingly redefined, and the process reiterates; otherwise, the plurality of top-ranking records is outputted.

Abstract: A method, system and computer program product for enhanced database and user interface incorporating predicted missing data. A reference record in a database comprising a plurality of values divided into a plurality of categories is received. A comparison computational operator for calculating similarity score is defined using criteria for determining a level of match between values of the reference record and a record compared thereto in each of the plurality of categories. Top ranked records according to the similarity score are selected, and for values included therein in a category of the plurality of categories, a frequency score comprising a count aggregated per similarity score of containing records is calculated and according thereto a set of missing values in the reference record is selected for enhancement of the reference record and/or presentation via a user interface.

Abstract: A method, system and computer program product for enhanced database and user interface incorporating predicted missing data. A reference record in a database comprising a plurality of values divided into a plurality of categories is received. A comparison computational operator for calculating similarity score is defined using criteria for determining a level of match between values of the reference record and a record compared thereto in each of the plurality of categories. Top ranked records according to the similarity score are selected, and for values included therein in a category of the plurality of categories, a frequency score comprising a count aggregated per similarity score of containing records is calculated and according thereto a set of missing values in the reference record is selected for enhancement of the reference record anchor presentation via a user interface.

Abstract: A method, system and computer program product for database search enhancement and interactive user interface therefor. Database records are ranked by a match score, calculated using a plurality of criteria for determining a match level between values of a record and a search query, and using a plurality of priority parameters for aggregating the match level determined. A plurality of top-ranking records is selected, and a diversity measure is calculated therefor, using at least one class label assigned to records therein. If a sufficiency condition is not met by the diversity measure, at least one reference set of records sharing a class label in common is extracted from the plurality of top-ranking records and analyzed for determining at least one modification to the search query in improvement of the diversity measure, the scoring computational operator is accordingly redefined, and the process reiterates; otherwise, the plurality of top-ranking records is outputted.

Abstract: A method of machine learning for generating a predictive model of a response characteristic based on historical data elements using a processor may include receiving historical data elements and historical values for the response characteristic related to uses of the historical data elements in web pages. A plurality of key-value pairs may be generated defining values of a plurality of predefined features representing properties of the historical data elements. Each of a plurality of n features may be represented by an axis in an n-dimensional space are extracted from the historical data elements. The extracted plurality of key-value pairs for each historical data element may be projected onto the n-dimensional space. The plurality of vectors may be input into a model generator to generate a predictive model predicting a value of the response characteristic for a new data element.