Abstract: Certain aspects of the present disclosure provide techniques for detecting sentences that are utterances by an agent indicating an intent to poach a customer. According to certain embodiments, a language model is trained using query sentences that are confirmed to be sentences used in poaching a customer, to identify semantically similar sentences in a corpus. These semantically similar sentences are then used as base sentences for comparison to sentences in a transcript. Sentences of the transcript that are found to be semantically similar to one or more base sentences are provided to a user device for review and confirmation that the similar sentence was generated by an agent in an attempt to poach a customer.

Abstract: A method including identifying first and second training data having first and second subsets of click-through information for a dataset. Identifying includes associating the first and second subsets first and second applications executing on first and second domains having divergent first and second ontologically defined groupings of entities. The method also includes storing, as first and second vector data structures, the first and second training data. The method also includes training, on the first and second vector data structures, first and second ARIMA machine learning models. The first trained ARIMA machine learning model is trained on the first domain and the second trained ARIMA machine learning model is trained on the second domain. The method also includes deploying the first and second trained ARIMA machine learning models.

Abstract: A method including receiving an input and embedding the input into a first data structure that defines first relationships among images and texts. The method also includes comparing the first data structure to an index including a second data structure that defines second relationships among pre-determined texts and pre-determined images. The method also includes returning a subset of images from the pre-determined images. The subset includes those images in the pre-determined images for which matches exist between the first relationships and the second relationships.

Abstract: A visual-based classification model influenced by text features as a result of the outputs of a text-based classification model is disclosed. A system receives one or more documents to be classified based on one or more visual features and provides the one or more documents to a student classification model, which is a visual-based classification model. The system also classifies, by the student classification model, the one or more documents into one or more document types based on one or more visual features. The one or more visual features are generated by the student classification model that is trained based on important text identified by a teacher classification model for the one or more document types, with the teacher classification model being a text-based classification model. Generating training data and training the student classification model based on the training data are also described.

Abstract: In a pipeline, data events generated by a producer application are temporally grouped by using a group identification tag. For each data event, data points are generated and uploaded to a storage and cache at each point of production and consumption. The storage allows a matching of data events between the production point and the consumption point, thereby ensuring that streaming parity is maintained. In cases of mismatch, the cache allows for detecting missing data events, i.e., identifying data events that were generated by an upstream producer application, but not consumed by a downstream consumer. While being agnostic to the transformations applied by the various applications in the pipeline, the embodiments disclosed herein keep track of the output data events and input data events and precisely identify the missing data events.

Abstract: A method converts a request received from a client application to a query by a serverless function operating on a virtual machine instance provisioned responsive to the request. The query is transmitted to an application database corresponding to the client application. A result is received from the application database. The result is converted into a response and the response is transmitted to the client application in response to the request. Display of the client application is updated based on response with the result.

Abstract: A method includes receiving a user input including natural language text. The method also includes generating modified inputs from the user input. The method also includes executing a machine learning model on the modified inputs to generate model outputs. The method also includes sampling the model outputs using a statistical sampling strategy to generate sampled model outputs. The method also includes clustering the sampled model outputs into clusters. The method also includes generating a confidence metric of the clusters. The method also includes routing, automatically in a computing system, the user input based on whether the confidence metric satisfies a threshold value.

Abstract: Systems and methods for assessment of user price sensitivity using a predictive model are disclosed.

Abstract: A method and system for generating anomaly-detection rules for communication protocols are provided. The method includes receiving communication data; constructing at least one N-gram from the received communication data; analyzing the at least one N-gram by comparing the constructed at least one N-gram with a repository of N-gram analyses to identify conditional probabilities of certain characteristics; and generating anomaly-detection rules based on the N-gram analysis.

Abstract: A method and system are provided for generating a combined prediction using an ensemble machine learning system. The prediction may be used in risk assessment for payroll processing. A data point is received as input to a multitude of trained models. Each model is trained from a respective data subset of a disparate data. A model prediction this generated by each of a multitude of machine learning models. For each respective trained model, a trust score is generated based on a data sparseness metric of the data point and a feature importance vector of the respective model. The model predictions and trust scores are received as input to a meta-model that was trained from the trust score and the model prediction of the multitude of trained models over the respective data subset of the disparate data. A combined prediction is generated using the trained meta-model.

Abstract: Customizable template-based user interfaces for digital CRM tools. For instance, the CRM tools are pre-loaded with customizable templates, which are rendered as interactive user interfaces that allow customization of the templates based on user needs. This customization therefore allows a user to rapidly, visually, and easily define a customer engagement pipeline and customer segments without any specialized programming knowledge. The user interfaces may dynamically show the flow of customers between different phases and provide recommended actions to further drive customer engagement. Additionally, the CRM tools are integrated with other on-line platforms such as social media sites and e-commerce sites.

Abstract: A method utilizes a framework for transaction categorization personalization. A transaction record is received. a baseline model is selected from a plurality of machine learning models. An account identifier, corresponding to the transaction record using the baseline model, is selected. The account identifier for the transaction record is presented.

Abstract: Certain aspects of the present disclosure provide techniques for managing a search engine based on search performance metrics. An example method generally includes dividing a set of search history data into a first subset of search history data and a second subset of search history data. The first subset of data is associated with interaction with search results, and the second subset of data is associated with non-interaction with search results. A first quality score is generated for searches in the first subset of data. A second quality score is generated for searches in the second subset of data based on different search intents identified for each temporally related group in the second subset of data. An overall quality score is generated for a search engine, and one or more actions with respect to the search engine are taken based on the overall quality score.

Abstract: Certain aspects of the present disclosure provide techniques for orchestrating a user experience using natural language input. A user experience is orchestrated within an ecosystem of features for which a plurality of corresponding tokens is defined. Natural language describing a desired user experience result is received by a user experience orchestrator. A sequence of tokens corresponding to operations belonging to an ecosystem of features which produce a correct result for the natural language input can be identified by a trained large language model and executed by the user experience orchestrator using a token operator. The output operations determined by the model to produce or be likely to produce the correct result based on the natural language input can be disambiguated, confirmed, and/or executed.

Abstract: A method performs personalized transaction categorization. A transaction record is received, by a server application. In a first stage, sparse raw features are extracted from a transaction record of a transaction and converted into a transaction vector including dense features. In a second stage, the transaction vector is classified into a customized chart of accounts using the dense features to generate adapter model output. The method further includes selecting, an account identifier, corresponding to the transaction record and to an account of the customized chart of accounts, using the adapter model output, and presenting the account identifier for the transaction record.

Abstract: This disclosure provides systems, methods and apparatuses for machine learning. In some implementations, a pre-processing system may identify one or more special features in an input dataset and may generate one or more pruned datasets, respectively, based on the identified special features. The pre-processing system maps each of the pruned datasets to one or more predictions and selectively removes the values associated with one or more special features from the input dataset based on the mapping. In some other implementations, a pre-processing system may map each of the feature sets in an input dataset to multiple predictions in accordance with multiple machine learning algorithms. The pre-processing system evaluates a performance of each of the mappings and selects one of the machine learning algorithms to be used to train the machine learning model based on the performance of each mapping.

Abstract: A method implements a dataset rank metric for measuring dataset relevance. Metadata is identified for a plurality of datasets. A graph structure is generated in storage. The graph structure includes a multitude of nodes connected by a multitude of edges. Each node of the multitude of nodes representing a respective dataset of a multitude of datasets, the multitude of edges connecting the multitude of nodes according to a data lineage determined from metadata of the multitude of datasets. A composite score is generated for each node of the graph. The computer processor iteratively processes the composite scores for the multitude of nodes of the graph to generate a dataset rank for each dataset. The multitude of datasets is presented in an interface, sorted according to the respective dataset rank of each dataset.

Abstract: A testing system and method for testing application code against various failure scenarios. The testing system and method generate a test workflow including test source code implementing a series of actions that affect an application component and or an infrastructure component included in application code. The testing system and method execute the test workflow to determine the performance of the application code during one or more failure scenarios caused by the series of actions included in the test workflow. Performance data generated by the test code is analyzed by a performance analysis service or method to identify limitations of the application code and build resiliency patterns that address the limitations and improve the performance of the application code.