Abstract: The present invention consists of a computer-implemented system and method for automatically analyzing and coding documents into content categories suitable for high cost, high yield settings where quality and efficiency of classification are essential. A prototypical example application field is legal document predictive coding for purposes of e-discovery and litigation (or litigation readiness) where the automated classification of documents as “responsive” or not must be (a) efficient, (b) accurate, and (c) defensible in court. Many text classification technologies exist but they focus on the relatively simple steps of using a training method on training data, producing a model and testing it on test data. They invariably do not address effectively and simultaneously key quality assurance requirements. The invention applies several data design and validation steps that ensure quality and removal of all possible sources of document classification error or deficiencies.

Abstract: A system and method for identifying critical positive and negative factors for the success of a research and development activity.

Abstract: The present invention addresses two ubiquitous and pressing problems of modern data analytics technology. Many modern pattern recognition technologies produce models with excellent predictivity but (a) they are “black boxes”, that is they are opaque to the user; (b) they are too large, and/or expensive to execute in less powerful computing platforms. The invention “opens up” a black box model by converting it to a compact and understandable model that is functionally equivalent. The invention also converts a predictive model into a functionally equivalent model into a form that can be implemented and deployed more easily or efficiently in practice. The benefits include: model understandability and defensibility of modeling. A particularly interesting application is that of understanding the decision making of humans, comparison of the behavior of a human or computerized decision process against another and use to enhance education and guideline compliance/adherence detection and improvement.

Abstract: Predictive modeling is an important class of data analytics with applications in numerous fields. Once a predictive model is built, validated, and applied on a set of objects, by a data analytics system (or even by manual modeling), consumers of the model information need assistance to navigate through the results. This is because both regression and classification models that output continuous values (eg, probability of belonging to a class) are often used to rank objects and then a thresholding of the ranked scores needs to be used to separate objects into a “positive” and a “negative” class. The choice of threshold greatly affects the true positive, false positive, true negative, and false negative results of the model's application. An ideal data analytics system should allow the user to understand the tradeoffs of different threshold values for different thresholds.

Abstract: Established methods for statistical “power-size” analysis for statistical modeling are geared toward statistical hypothesis testing, and have serious shortcomings in modern complex predictive and causal modeling applications where the determination of sample size is affected by parameters not addressed by the standard statistical power-size analysis. The present invention provides a method and computer-implemented system for determining sufficient sample size for training predictive or causal models for a given application field or distribution type and desired performance level taking into account the critical factors that affect the needed sample size. The invention can be applied to practically any field where predictive modeling or causal modeling are desired.

Abstract: The present invention addresses two ubiquitous and pressing problems of modern data analytics technology. Many modern pattern recognition technologies produce models with excellent predictivity but (a) they are “black boxes”, that is they are opaque to the user; (b) they are too large, and/or expensive to execute in less powerful computing platforms. The invention “opens up” a black box model by converting it to a compact and understandable model that is functionally equivalent. The invention also converts a predictive model into a functionally equivalent model into a form that can be implemented and deployed more easily or efficiently in practice. The benefits include: model understandability and defensibility of modeling. A particularly interesting application is that of understanding the decision making of humans, comparison of the behavior of a human or computerized decision process against another and use to enhance education and guideline compliance/adherence detection and improvement.

Abstract: Predictive modeling is an important class of data analytics with applications in numerous fields. Once a predictive model is built, validated, and applied on a set of objects, by a data analytics system (or even by manual modeling), consumers of the model information need assistance to navigate through the results. This is because both regression and classification models that output continuous values (eg, probability of belonging to a class) are often used to rank objects and then a thresholding of the ranked scores needs to be used to separate objects into a “positive” and a “negative” class. The choice of threshold greatly affects the true positive, false positive, true negative, and false negative results of the model's application. An ideal data analytics system should allow the user to understand the tradeoffs of different threshold values for different thresholds.

Abstract: The present invention consists of a computer-implemented system and method for automatically analyzing and coding documents into content categories suitable for high cost, high yield settings where quality and efficiency of classification are essential. A prototypical example application field is legal document predictive coding for purposes of e-discovery and litigation (or litigation readiness) where the automated classification of documents as “responsive” or not must be (a) efficient, (b) accurate, and (c) defensible in court. Many text classification technologies exist but they focus on the relatively simple steps of using a training method on training data, producing a model and testing it on test data. They invariably do not address effectively and simultaneously key quality assurance requirements. The invention applies several data design and validation steps that ensure quality and removal of all possible sources of document classification error or deficiencies.

Abstract: The learning method taught in this patent document is significantly different from previous methods for automatic classification of citations that are labor intensive and subject to human bias and error. The present invention automatically generates and avoids these limitations. A set of operational definitions and features uniquely suited to the scientific literature is disclosed along with their use with a learning method that is capable of analyzing the textual content of articles along with bibliometric data to accurately classify instrumental citations.

Abstract: A computerized process to predict citation counts of articles using only information available before or at the time of publication. The process involves obtaining a set of articles and extracting a set of features containing information about the article, author, and bibliometric data. The extracted features are converted into a format suitable for analysis, and models are constructed using a pattern recognition process. The constructed models are applied to a related article that was not included in the original article set for model construction. Features are extracted from the article of interest, and the models provide a prediction of whether a given number of citations will be received by the article.

Abstract: The learning method taught in this patent document is significantly different from previous methods for automatic classification of citations that are labor intensive and subject to human bias and error. The present invention automatically generates and avoids these limitations. A set of operational definitions and features uniquely suited to the scientific literature is disclosed along with their use with a learning method that is capable of analyzing the textual content of articles along with bibliometric data to accurately classify instrumental citations.

Abstract: A computerized process to predict citation counts of articles comprising the steps of receiving an article through an input, obtaining, through the input, a selected set of articles exclusive of the article, storing in a memory the set of articles and the article, extracting through a computer processor an article feature from each article in the stored set of articles, constructing models through the computer processor using a pattern recognition process and the article feature, selecting, through the processor, a best model, predicting by application of the best model to the article by the processor a future citation count of the article, outputting, the article comprising the future citation count and controlling through a publication controller unit, distribution of the article.