Abstract: An anomaly detection system using machine learning to generate predicted survey scores for a given duration and a given metric based on historic survey score data. The system compares a predicted survey score to the actual survey score and identifies anomalous actual survey scores. The anomaly detection system trains a plurality of survey score prediction models using historic survey score data. Each survey score prediction model is based on a specific survey score metric and a specific duration. The survey score prediction models generate expected survey score results for the given duration and the given metric. Based on the user-determined filtering and tolerances, the system determines if the actual survey score result is anomalous. The system generates reports for the detected anomalies and continually updates the survey score prediction models with newly obtained actual survey results, thereby improving the anomaly detection accuracy over time.

Abstract: Detecting fraudulent activity can be a complex, manual process. In this paper, we adapt statistical properties of count data in a novel algorithm to uncover records exhibiting high risk for fraud. Our method identifies shelves, partitioning data under the counts using a Student's t-distribution. We apply this methodology on a univariate dataset including cumulative results from phone calls to a customer service center. Additionally, we extend this technique to multivariate data, illustrating that the same method is applicable to both univariate and multivariate data.

Abstract: Detecting fraudulent activity can be a complex, manual process. In this paper, we adapt statistical properties of count data in a novel algorithm to uncover records exhibiting high risk for fraud. Our method identifies shelves, partitioning data under the counts using a Student's t-distribution. We apply this methodology on a univariate dataset including cumulative results from phone calls to a customer service center. Additionally, we extend this technique to multivariate data, illustrating that the same method is applicable to both univariate and multivariate data.

Abstract: A scalable system provides automated conversation review that can identify potential miscommunications. The system may provide suggested actions to fix errors in intelligent virtual assistant (IVA) understanding, may prioritize areas of language model repair, and may automate the review of conversations. By the use of an automated system for conversation review, problematic interactions can be surfaced without exposing the entire set of conversation logs to human reviewers, thereby minimizing privacy invasion. A scalable system processes conversations and autonomously marks the interactions where the IVA is misunderstanding the user.

Abstract: A conversation user interface enables patients to better understand their healthcare by integrating diagnosis, treatment, medication management, and payment, through a system that uses a virtual assistant to engage in conversation with the patient. The conversation user interface conveys a visual representation of a conversation between the virtual assistant and the patient. An identity of the patient, including preferences and medical records, is maintained throughout all interactions so that each aspect of this integrated system has access to the same information. The conversation user interface presents allows the patient to interact with the virtual assistant using natural language commands to receive information and complete task related to his or her healthcare.

Abstract: Features, libraries, and techniques are provided herein for determining the kinds of relational language that are present. Applying audio, emojis, and sentiment shifts as features may be used to determine whether the customer is providing backstory, whether there is ranting, etc. Textual features may be considered, as well as audio features may be considered.

Abstract: Techniques for interacting with a portion of a content item through a virtual assistant are described herein. The techniques may include identifying a portion of a content item that is relevant to user input and causing an action to be performed related to the portion of the content item. The action may include, for example, displaying the portion of the content item on a smart device in a displayable format that is adapted to a display characteristic of the smart device, performing a task for a user that satisfies the user input, and so on.

Abstract: Techniques and architectures for implementing a team of virtual assistants are described herein. The team may include multiple virtual assistants that are configured with different characteristics, such as different functionality, base language models, levels of training, visual appearances, personalities, and so on. The characteristics of the virtual assistants may be configured by trainers, end-users, and/or a virtual assistant service. The virtual assistants may be presented to end-users in conversation user interfaces to perform different tasks for the users in a conversational manner. The different virtual assistants may adapt to different contexts. The virtual assistants may additionally, or alternatively, interact with each other to carry out tasks for the users, which may be illustrated in conversation user interfaces.

Abstract: Systems and apparatus for sensing and video capture include at least one camera with an optical sensor that captures video image data of a first sampling rate. An auxiliary sensor captures auxiliary data at a second sample rate. A processor is communicatively connected to the optical sensor and auxiliary sensor. The processor transmits video image data captured at the first sample rate auxiliary sensor data captured at the second sampling rate across a data connection to a centralized computer that receives the video image data and the auxiliary sensor data and operate to present the video image data and the auxiliary sensor data on a graphical display.

Abstract: Real-time speech analytics (RTSA) provides maintaining real-time speech conditions, rules, and triggers, and real-time actions and alerts to take. A call between a user and an agent is received at an agent computing device. The call is monitored to detect in the call one of the real-time speech conditions, rules, and triggers. Based on the detection, at least one real-time action and/or alert is initiated.

Abstract: A set of documents related to a particular topic, industry, or entity are received. Sentences are extract from each document. The sentences are grouped into tuples of one, two, or three consecutive sentences (i.e., short text sequences). The sentence tuples are clustered based on vector representations of the sentences. For each cluster, a set of tuples that best represents or best fits the cluster is selected. These sentence tuples are fed to an ontology to determine ontological entities associated with each tuple. These determined ontological entities are associated with the clusters corresponding to each tuple. The sentence tuples associated with each cluster are labeled based on the ontological entities associated with the cluster. The labeled sentence tuples may then be used for a variety of purposes such as training a model to determine the topic of short text sequences.

Abstract: A system and method using blockchain for monitoring and tracking service provider involvement in a transaction on behalf of a customer company. In the system and method, session information related to the transactions are encrypted using an encryption key specific to a company on whose behalf the service provider is acting. The encrypted action is signed the with a private key of a public/private key pair. The signed, encrypted action record is placed on the blockchain, which can later be accessed to review the actions on behalf of the specific company.

Abstract: A method for workforce scheduling by a computer system is provided. The method includes receiving a first workforce schedule describing initial assignments of a plurality of workers to a plurality of shifts, each shift comprising one or more work activities, each work activity comprising an activity and a time interval, and storing the first workforce schedule in a memory. The method also includes receiving a cell size associated with each activity, and determining a quantity of workers in each work activity associated with each activity in the first workforce schedule. The method further includes determining cell size violations by dividing the quantity of workers assigned to each work activity by the cell size for the activity associated with the work activity. The method also includes modifying the first workforce schedule to minimize cell size violations, resulting in a second workforce schedule, and storing the second workforce schedule in the memory.

Abstract: The present invention allows text analysis and routing of an outgoing message. The system intercepts outgoing messages for analysis by a TAS software module. The module assigns an analytical score to the message, then compares the score to a threshold. If the score is below the threshold, the message is transmitted to its ultimate destination. If not, the message may be routed for correction by the message's composer or quality assurance staff. After such correction, the message new analytical score is generated and compared, and, if necessary, the process repeats again.

Abstract: A system and method for updating computerized language models is provided that automatically adds or deletes terms from the language model to capture trending events or products, while maximizing computer efficiencies by deleting terms that are no longer trending and use of knowledge bases, machine learning model training and evaluation corpora, analysis tools and databases.

Abstract: An IVR and chatbot, or other system, employing a language model, the language model resulting from a method and computer product encoding the method is available for preparing a domain or subdomain specific glossary. The method included using probabilities, word context, common terminology and different terminology to identify domain and subdomain specific language and a related glossary updated according to the method.

Abstract: The present invention is a method and system for automatedly producing at least one desktop analytics trigger. Upon receiving at least one type of data input, the system analyzes the data input and produces at least one desktop analytics trigger based on the results of the analysis of the data input. The data input can include data on the programs, applications, or information a user utilizes during a task, to allow use of desktop process analytics. This process may be used to either generate a new desktop analytics trigger or update an existing desktop analytics trigger.

Abstract: Systems and methods for incorporating intelligent virtual assistants into advertisements on social networking platforms are provided. When a user interacts with a content item, an intelligent virtual assistant is selected and put into contact with the user. The intelligent virtual assistant is provided with a context that includes information about the user in the social networking platform, information about the user in a customer relationship management platform, and information about the product, service, or entity associated with the content item. The context allows the intelligent virtual assistant to converse with the user in a way that feels natural and relevant to the user and allows the intelligent virtual assistant to answer any questions about the product, service, or entity associated with the content item.

Abstract: A computerized system for deriving expression of intent from recorded speech includes: a text classification module comparing a transcription of recorded speech against a text classifier to generate a first set of representations of potential intents; a phonetics classification module comparing a phonetic transcription of the recorded speech against a phonetics classifier to generate a second set of representations; an audio classification module comparing an audio version of the recorded speech with an audio classifier to generate a third set of representations; and a discriminator module for receiving the first, second and third sets of the representations of potential intents and generating one derived expression of intent by processing the first, second and third sets together; where at least two of the text classification module, the phonetics classification module, and the audio classification module are asynchronous processes from one another.

Abstract: Various embodiments are described for searching and retrieving documents based on a natural language input. A computer-implemented natural language processor electronically receives a natural language input phrase from an interface device. The natural language processor attributes a concept to the phrase with the natural language processor. The natural language processor searches a database for a set of documents to identify one or more documents associated with the attributed concept to be included in a response to the natural language input phrase. The natural language processor maintains the concepts during an interactive session with the natural language processor. The natural language processor resolves ambiguous input patterns in the natural language input phrase with the natural language processor. The natural language processor includes a processor, a memory and/or storage component, and an input/output device.