US 11,055,320 B2
System for mapping employees sentiments and a method thereof
Udit Chandna, Delhi (IN); Sandeep Kishore, Fremont, CA (US); Kapil Bharti, San Jose, CA (US); and Ullas Balan Nambiar, Bangalore (IN)
Assigned to Zensar Technologies Ltd., Maharashtra (IN)
Filed by Zensar Technologies Ltd., Maharashtra (IN)
Filed on Dec. 5, 2018, as Appl. No. 16/210,828.
Claims priority of application No. 201721044011 (IN), filed on Dec. 7, 2017.
Prior Publication US 2019/0179834 A1, Jun. 13, 2019
Int. Cl. G06F 16/28 (2019.01); G06N 20/00 (2019.01); G06Q 10/06 (2012.01); G06F 16/951 (2019.01)
CPC G06F 16/285 (2019.01) [G06F 16/951 (2019.01); G06N 20/00 (2019.01); G06Q 10/06393 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A system (100) for mapping employees' sentiments, said system (100) comprising:
a processor; and
a memory storing a database (15) configured to store a pre-determined group of themes and at least one pre-determined keyword corresponding to each of said themes, a pre-determined weightage corresponding to each of said keywords, a pre-determined set of scoring rules, a pre-determined list of stop words, and a pre-determined list of keywords, and wherein the memory is coupled to the processor, and wherein the processor is capable of executing a set of instructions stored in the memory, and wherein the set of instructions comprising:
receiving an employee input in an audio format and thereby converting the employee input received in the audio format to a textual format by using an analogue to digital converter;
generating tokens based on said employee input;
filtering out words, based on said pre-determined list of stop words and said tokens;
extracting keywords, by using a lexical analysis technique, from said pre-determined list of keywords based on filtered words, wherein said pre-determined list of keywords is periodically updated with new keywords using a machine learning technique implemented in a manner such that the system (100) self-learns;
crawling through said pre-determined keyword corresponding to each of said themes using said extracted keywords to identify at least one keyword;
identifying said pre-determined keywords corresponding to each of said themes based on said extracted keywords;
searching through said pre-determined group of themes using said identified keywords, further configured to map said identified keyword to at least one of said pre-determined group of themes;
computing a weighted score for each of said identified keyword corresponding to said mapped group of themes using said pre-determined weightage, further configured to compute a quantitative score for each of said mapped group of themes based on said weighted score and said pre-determined set of scoring rules;
analyzing employee sentiments, based on said quantitative score; and
computing
an employee satisfaction score based on a relationship between said employee sentiments, employee feedback and employee performance related to said employee, and
a cumulative employee satisfaction score based on a percentage of employees impacted by actions taken, time elapsed between the employee feedback and the actions taken, and time spent on performing the actions.