Adaptive and reusable processing of retroactive sequences for automated predictions

Abstract

Systems and methods for predicting the outcome of a business entity are presented. In embodiments, a system may receive explicit data reporting or indicating activities of a business entity, and other data from which information regarding the activities or level of operations of the entity may be inferred. Using one or more data processors, the system may generate inferred data regarding the business entity from a selected portion of the other data, and use at least some of the explicit data and the inferred data to determine which one of a series of defined sequential active states of development the entity currently is in. The system may further, using the result of the determination as the current state of the business, predict a final stage of the business entity, and a probability of evolving to that final stage from the current state. Other embodiments may be disclosed or claimed.

What is claimed: 
     
         1 . A method of adaptive and reusable processing of retroactive sequences for automated predictions, comprising:
 receiving explicit data reporting or indicating activities of a plurality of business entities;   receiving other data from which information regarding the activities or level of operations of the business entities may be inferred;   using one or more data processors:
 generating inferred data regarding the business entities from a selected portion of the other data; 
 processing the explicit data and the inferred data to generate a database of at least some of the business entities, entries in the database comprising at least one of partial timelines or sets of sequences of events of the business entities; 
 processing the database to infer lifecycles for the business entities, a lifecycle comprising a final state, and a sequence of active states a business entity passes through prior to the final state; 
 outputting the inferred lifecycles to a user. 
   
     
     
         2 . The method of  claim 1 , wherein the final state is one of closed, acquired, initial public offering and acquired to obtain employees. 
     
     
         3 . The method of  claim 1 , wherein the sequences of active states include one or more of concept, seed, product development, market development and steady status. 
     
     
         4 . The method of  claim 3 , wherein the sequences of active states conform to a defined series of hidden active states that a business entity may go through, and wherein in the defined series of active states, multiple actual events of the companies are subsumed by at least one of the active states. 
     
     
         5 . The method of  claim 1 , wherein the final state of the business entities is observed. 
     
     
         6 . The method of  claim 1 , wherein processing the database to infer lifecycles includes determining, for each of the sequence of active states in the lifecycle, expected values for a set of scores indicative of that active state. 
     
     
         7 . The method of  claim 6 , further comprising determining a vector that expresses a probability of transitioning from that active state to the final state of the lifecycle. 
     
     
         8 . The method of  claim 1 , wherein processing the database to infer lifecycles includes associating a Complex Hidden Markov Model (CHMM) with each company. 
     
     
         9 . The method of  claim 1 , wherein processing the database to infer lifecycles includes reconstructing sequences of potential active states in multiple passes, the passes separated in time from each other, and using different sources of information. 
     
     
         10 . The method of  claim 1 , wherein lifecycles are generated for groups of companies and for individual companies. 
     
     
         11 . A memory having instructions stored thereon that, in response to execution by a processor, cause the processor to perform operations comprising:
 uploading digital input data for a person associated with a business entity from at least one data source;   provisioning a plurality of feature synthesizers to synthesize feature data from the input data for each attribute of a set of attributes to form at least one dataset of character feature data associated with the person;   provisioning a plurality of specialized model builder components, each specialized model builder component configured to build a prediction model for a corresponding one of the attributes;   training at least some of the specialized model builder components based at least in part on the synthesized feature data to form respective prediction models for the attributes, respectively;   separately applying each of the trained prediction models to at least a portion of the character feature data associated with the person to generate individual vectors corresponding to attributes, respectively, wherein the individual vectors form a feature vector;   identifying a plurality of additional feature vectors, each of which includes individual vectors corresponding to the attributes, respectively, for a different person associated with the business entity;   combine the individual vectors of the feature vector with the individual feature vectors of the additional feature vectors, respectively, to form an aggregate score vector; and   reporting information about the aggregate score vector via a display interface.   
     
     
         12 . The memory of  claim 11 , wherein reporting the information about the aggregate score vector via the display interface includes generating and displaying a visualization including graphical elements for individual vectors of the aggregate score vector, respectively, wherein areas of the graphical elements are sized to represent data taken from the corresponding individual vector of the aggregate score vector. 
     
     
         13 . The memory of  claim 11 , wherein the operations further comprise:
 repeating the uploading and synthesizing steps to acquire new data;   separately applying each of the trained prediction models to the new data to form a new individual vector for each one of the attributes for the person; and   comparing the new individual vectors to the individual vectors of the feature vector to determine mutability of at least one of the attributes.   
     
     
         14 . The memory of  claim 11 , wherein the operations further comprise:
 comparing the aggregate score vector to a predetermined preferred distribution; and   reporting a result of the comparison via the display interface.   
     
     
         15 . The memory of  claim 11 , wherein the operations further comprise:
 identifying attributes of the aggregate score vector that deviate from a predetermined preferred distribution; and   reporting which attributes of the team aggregate score vector deviate from the predetermined preferred distribution.   
     
     
         16 . The memory of  claim 11 , wherein the operations further comprise:
 in forming the business entity score, separately for each attribute, combining the corresponding individual scores based on a predetermined Boolean operator selected to reflect a desired composition of the business entity with regard to the corresponding attribute.   
     
     
         17 . The memory of  claim 16 , wherein a Boolean AND operator is applied to combine the corresponding individual scores for an attribute that is required to be true according to a predetermined preferred distribution. 
     
     
         18 . The memory of  claim 16 , wherein a Boolean XOR operator is applied to combine the corresponding individual scores for an attribute that is required to be true according to a predetermined preferred distribution. 
     
     
         19 . The memory of  claim 11 , wherein the operations further comprise:
 assessing a mutability metric for at least one of the attribute for a team member, based on changes in the corresponding attribute score over time, wherein the mutability metric comprises a numeric value and a sign indicating a direction of change of the corresponding scores.   
     
     
         20 . The memory of  claim 19 , wherein the operations further comprise including the mutability metric in the corresponding attribute score in the first feature vector.