Morgan Stanley Services Group Inc.

Cheryl Fernandes

Mehak Mehta

Aratrika Sarkar

Melissa Kagaju

William R Schnieders

Rahul Suresh

Vijay Kumar Butte

Jayati Tripathi

Sandeep Ramesh

Alisha Singh

Labdhi Shah

Naina Bharadwaj

Dipesh Singh

Himani Garg

Sushil Ninawe

Alex Gavrilov

A method of implementing a regulation and directive to code platform comprises automatically pulling updates of a new regulation document from policy regulation sources, chunking the regulation document into smaller segments, generating embeddings of the segments of the using an LLM, classifying and extracting metadata from the new regulation document based on the classification, storing the embeddings of the segments with metadata tags in a retrieval automated generation (RAG) database, storing the extracted metadata extracted in an enterprise policy repository, extracting actionable items from the segments into a code form based on a prompt using an LLM, extracting template artifacts from the segments based on another prompt using an LLM, storing the extracted actionable items and template artifacts in the enterprise policy repository, and generating a central policy user interface through which users can access, review and update all regulation and directive information including the actionable items and template artifacts.

Regulations and directives to code platform

Systems and methods for detecting human impersonations on the internet, comprising receiving, by an ingestion layer of a detection system, a plurality of web data; filtering, by a processing engine of the detection system, the plurality of web data to identify relevant websites; extracting, via a scoring engine of the detection system, categories of data of a website of the relevant websites; assigning a score, via the scoring engine, for each category of the categories of data; and determining a risk score for the website, via the scoring engine, based on the score for each category.

Automated identification of internet human impersonators

A device, system and method for analyzing audio speech signals to detect fraudulent calls to a contact center comprising splitting an audio recording of a call in real-time into a foreground speech signal attributed to a main speaker and a background audio signal, extracting audio features from the foreground speech signal and background audio signal, inputting the extracted audio features into an ensemble model comprising multiple different machine learning models co-trained to cumulatively detect fraud, wherein the multiple different machine learning models include: a speaker audio model to detect audio speech anomalies, a speaker intent model to classify intent of the main speaker, a synthetic voice detection model to identify a non-human entity, and a prosody model to detect voice intonation of the main speaker. A prediction may be output, by the ensemble model, indicating whether the call is fraudulent.

Audio speech signal analysis for fraud detection

A device, system and method for analyzing audio speech signals to detect fraudulent calls to a contact center comprising splitting an audio recording of a call in real-time into a foreground speech signal attributed to a main speaker and a background audio signal, extracting audio features from the foreground speech signal and background audio signal, inputting the extracted audio features into an ensemble model comprising multiple different machine learning models co-trained to cumulatively detect fraud, wherein the multiple different machine learning models include: a speaker audio model to detect audio speech anomalies, a speaker intent model to classify intent of the main speaker, and a prosody model to detect voice intonation of the main speaker. A prediction may be output, by the ensemble model, indicating whether the call is fraudulent.

Growing community of inventors

Plainsboro, NJ, United States of America

Cheryl Fernandes