Fmr Corp.

Tredence Inc.

Aravind Chandramouli

Ankush Chopra

Abhishek Desai

Sunil Kumar

Nitin Hardeniya

Shubham Sharma

Manish Gupta

Rajib Biswas

Shiven Purohit

Vrushali Karnik

Paarth Ajay Shrivastava

Mohammed Sameer Khan

Vijay Goel

Divik Saxena

Siddharth Shukla

Ashutosh Rajesh Kothiwala

Ankit Patil

Soumendra Mohanty

Javeed Bani

Madhuri Gupta

Shubham Pandey

Embodiments include a method. The method includes receiving, by the computer system, a first input. The method includes generating, by at least using a language detection model, a set of predicted languages. The method includes generating, using different translation models, a different language translation of the first input for each language in the set of predicted languages. The method includes generating, using a first question-answer model, a set of answers. The method includes performing a distribution-based correction, by at least using a distribution source and at least one answer included in the set of answers, to determine the attribute represented by the arbitrary values included in the first input. The method includes causing a device to present a recommendation based at least in part on the attribute represented by the arbitrary values included in the first input.

Attribute extraction from text sources

An analytics system receives, from a client computing device, a request to generate a presentation. The analytics system accesses one or more feedback datasets of feedback data. The feedback data comprises unstructured data available from multiple data stores. The analytics system generates, for each feedback dataset, a respective feedback text and a respective sentiment score indicating a degree of negativity associated with the respective feedback text. For a combination of a plurality of generated feedback texts, the analytics system selects a set of themes based at least on a plurality of generated sentiment scores. Each sentiment score of the plurality of generated sentiment scores is associated with one of the plurality of generated feedback texts. The analytics system generates a presentation file that indicates the set of themes. The analytics system causes the presentation file to be transmitted to the client computing device.

Multi-channel feedback analytics for presentation generation

A computerized method of representing customer interactions with an organization includes: receiving, by a computing device, customer web interaction data segments and customer conversation data segments; pre-processing the customer conversation data segments to remove specified types of information; scoring each of the pre-processed customer conversation data segments; pre-processing the customer web data interaction segments; extracting from the pre-processed customer web interaction data segments tokens; combining the pre-processed customer conversation data segments and the pre-processed customer web interaction data segments into a customer data set; parsing the customer data set into one or more windows; assigning, for each window, pre-trained weights to each of the tokens in each window; assigning a transaction theme to each window based on the tokens in each window; and generating, based on the transaction themes, a ranked list of topic keywords reflecting the customer web interaction data segments and customer conversation data segments.

Computer systems and methods for efficient query resolution by customer representatives

A computerized method is provided for automatically identifying a set of historical issues derived from historical customer interactions with an enterprise. The method includes collecting text-based data corresponding to the historical customer interactions, extracting customer queries from the text-based data, and normalizing and filtering the customer queries to generate the set of historical issues of the customer queries. The method also includes assigning the historical issues to one or more clusters that capture variances among the historical issues. The method further includes generating a legitimacy score for each historical issue and ranking the set of historical issues in accordance with their corresponding legitimacy scores. The method can further include identifying one or more issues in a transcript of unstructured text using the set of historical issues.

Automatic identification of issues in text-based transcripts

A computerized method of populating one or more structured databases includes performing, by a computing device, the steps of: receiving customer message data from one or more data sources; extracting, from the customer message data, data sets representative of a set of customer questions; pre-processing, the data sets representative of the set of customer questions using one or more filters, thereby producing pre-processed data sets representative of customer questions; extracting, from the pre-processed data sets representative of customer questions, a set of customer expression data sets; grouping, the customer expression data sets into a set of clusters, each cluster representing one customer intent data set, each customer intent data set corresponding to one or more customer expression data sets; and storing, the customer intent data sets and the customer expression data sets in the structured database(s), the structured database(s) in electronic communication with the computing device.

Computer systems and methods to discover questions and answers from conversations

Methods and apparatuses are described for analyzing unstructured computer text for identification and classification of complaint-specific interactions. A computer data stores unstructured text. A server computing device splits the unstructured text into phrases of words. The server generates a set of tokens from each phrase and removes tokens that are stopwords. The server generates a normalized sentiment score for each set of tokens. The server uses a rules-based classification engine to generate a rules-based complaint score for each set of tokens. The server uses an artificial intelligence machine learning model to generate a model-based complaint score for each set of tokens. The server determines determine whether each set of tokens corresponds to a complaint-specific interaction based upon the rules-based complaint score and the model-based complaint score.

Classifying unstructured computer text for complaint-specific interactions using rules-based and machine learning modeling

Methods and apparatuses are described for analyzing unstructured computer text for domain-specific stopword identification and removal. A computer data store stores unstructured text. A server computing device splits the unstructured text into phrases and generates tokens from the phrases. The server computing device generates a set of bootstrap keywords using the tokens. An artificial intelligence neural network executing on the server computing device generates a stopword training model. The server computing device generates a first set of candidate stopwords using the bootstrap keywords and the stopword training model. The server computing device generates regular expressions using the bootstrap keywords, and generates a second set of candidate stopwords using the regular expressions. The server computing device stores the candidate stopwords in the data store, and removes stopwords from the unstructured text using the data store.

Domain-specific stopword removal from unstructured computer text using a neural network

Methods and apparatuses are described for analyzing unstructured computer text for theme generation to determine sentiment. A computer store stores unstructured text that is delimited, a searched phrases log, and a phrase click log. A computer server extracts phrases from the unstructured delimited text by splitting each line of the unstructured delimited text into one or more phrases. The computer server generates tokens from the unstructured delimited text, where the tokens comprise segments of the unstructured delimited text. The computer server determines one or more themes present in the unstructured delimited text.

Growing community of inventors

Bengaluru, India

Aravind Chandramouli