Abstract: One embodiment provides a method of using a computing device to analyze and generate a portfolio action strategy for clients including receiving a portfolio with stock information on one or more stocks held in the portfolio. The computing device receives a guideline for actions for the one or more stocks in the portfolio. The computing device further receives a historical set of data for a plurality of stocks. The computing device additionally trains a machine learning model that combines, analyzes and clusters stocks in the plurality of stocks by volatility, segment, stability, and volume. The computing device receives a current state of the one or more stocks held in the portfolio. The computing device further analyzes, utilizing the trained machine learning model, the stock information and a current state of the one or more stocks held in the portfolio to create a portfolio action strategy for the portfolio.

Abstract: An approach to identifying architectures of machine learning models meeting a user defined constraint. The approach can receive input associated with evaluating machine learning models from a user. The approach can determine acceptable architectural templates to evaluate the machine learning models based on the input and determine a list of architectures and metrics based on a calculation of maximum neural network sizes of the acceptable architectural templates not exceeding the constraint. The approach can send the list of architectures and metrics to the user for selection.

Abstract: A method, computer system, and a computer program product for dynamic workflow adjustment is provided. The present invention may include generating a digital workflow twin, wherein the digital workflow twin is a digital twin of an original network. The present invention may include identifying one or more potential impact regions within the digital workflow twin. The present invention may include determining whether to propose one or more recommendations to a user. The present invention may include converting at least one of the one or more recommendations into at least one new rule. The present invention may include generating a new network workflow, wherein the new network workflow is the original network with the at least one new rule integrated.

Abstract: An embodiment for improved predicting and mitigating of controversial language in digital communications. The embodiment may detect language within a candidate communication associated with a user and a corresponding user network. The embodiment may interpret the detected language to identify a series of predetermined controversial traits. The embodiment may calculate an impact risk score for the candidate communication by using a graph neural network to leverage historical data within an accessible knowledge base, wherein the impact risk score is based on risk associated with the detected language and the corresponding user network. The embodiment may, in response to detecting the calculated impact risk score for the candidate communication is above a predetermined threshold value, automatically identify suitable replacement language. The embodiment may generate and output to the user one or more substitute communications including the identified suitable replacement language.

Abstract: Managing licenses is provided. Evidence required for compliance with conditions set forth in a license is received from a client device of a designated user via a network. An analysis of the evidence is performed using machine learning based on a set of criteria to determine whether the evidence is valid. It is determined, using machine learning, whether the evidence is valid based on the analysis of the evidence. The evidence required for the compliance with the conditions set forth in the license is sent to a regulatory agency via the network to maintain the license for an entity in response to the determining that the evidence is valid based on the analysis of the evidence.

Abstract: A method, system, and computer program product are configured to: receive an input machine learning model being developed for deployment in a production environment; create a feature vector based on metadata of the input machine learning model and a data source used by the input machine learning model; determine a risk tier of the input machine learning model by classifying the input machine learning model into one of plural predefined tiers using the feature vector with a model-tiering machine learning model; and provide validation information to an evaluator wherein the validation information is based on the risk tier and is used to validate the input machine learning model in accordance with the risk tier.

Abstract: A system for initiating web containers and isolating network traffic to web containers is described. The system initiates and uses web containers for associated web-based resources, such as domain addresses owned by a single entity. When a web container for a domain address has not been initiated, the system determines an owner of the domain address and any domain addresses associated with the owner and initiates a web container for the domain address and the associated domains. When a user navigates to the domain addresses associated with the determined owner, the system isolates the network activity to the initiated web container.

Abstract: The present disclosure provides for crop yield estimation by identifying, via image processing, a field in which a crop is grown; identifying a plurality of regions within the field; identifying, by processing growth metrics via a model, a plurality of data collection points in the plurality of regions, wherein a given data collection point of the plurality of data collection points within a given region of the plurality of regions is identified by multivariate analysis as representative of growing conditions in the given region; receiving in-field data linked to the data collection points of the plurality; and predicting a yield for the crop in the field based on the in-field data.

Abstract: In a first aspect of the invention, there is a computer-implemented method including: training, by a computing device, a machine learning (ML) model on a financial data set comprising data on financial behavior applicable to repaying debts; entering, by a computing device, an interaction with a user regarding renegotiating a debt; and generating, by the computing device, a proposed payment plan for the debt, based on user data of the user, and using the ML model trained on the financial data set.

Abstract: In an approach to composing a machine learning model for complex data sources, a computer receives data and associated metadata corresponding to a machine learning task from a user. A computer determines a task context and a problem domain. A computer identifies the machine learning task. A computer evaluates a match between the problem domain and one or more pre-compiled models. A computer selects at least two of the one or more pre-compiled models. A computer generates one or more multimodal model combinations with the selected at least two of the one or more pre-compiled models. A computer executes the multimodal model combinations with the data and associated metadata. A computer displays the results of the executed one or more multimodal model combinations to the user. A computer determines whether a level of error associated with the results is acceptable to the user based on a response from the user.

Abstract: In response to identifying one or more applications on a computing device, terms and conditions associated with the one or more applications are identified. A knowledge base based on the identified terms and conditions is generated. The knowledge base includes a determination of sensitive information of a user required by each of the identified one or more applications. Secondary associations of the identified one or more applications are identified from a plurality of external data sources with information related to the identified one or more applications. In response to identifying the secondary associations, it is determined whether sensitive information about the user is found in one or more of the identified secondary associations. In response to determining that sensitive information about the user is found, an advisory is transmitted to the user.

Abstract: Chatbot logs can be received. The chatbot logs can be clustered or classified into clusters of potential intents. Trending topics can be detected by analyzing at least social media content. Semantic similarities between the potential intents and the trending topics can be computed. Based on the computed semantic similarities, at least one of the potential intents to at least one of the trending topics can be correlated. Training of the chatbot using the chatbot logs associated with the correlated potential intent can be triggered.

Abstract: A system may include a memory and a processor in communication therewith. The processor may be configured to perform operations including obtaining an egocentric network with a center node for an entity and receiving data about the entity. The operations may include predicting, with the data, a node array of the egocentric network including a first node to remain, a second node to disappear, and a third node to join the network. The operations may include prognosticating, with the data, a connection array of the egocentric network including a first link connecting the first node to the center node, a second link disappearing from the egocentric network, and a third link connecting the third node to the center node. The operations may include updating the egocentric network to reflect the node and connection arrays, generating an output based on the egocentric network, and displaying the output to a user.

Abstract: An example operation may include one or more of storing existing chaincode relationships of a group of blockchain peers within a blockchain network, identifying a new chaincode to implement for one or more blockchain peers among the group of blockchain peers of the blockchain network based on the existing chaincode relationships among the group of blockchain peers, and transmitting a message to the one or more blockchain peers with a suggestion to implement the new chaincode.

Abstract: A topic for building a new intent on which to train a chatbot can be received. A database of chatbot training data can be searched for a candidate intent having meta-knowledge similar to the received topic. Utterances associated with the candidate intent can be extracted. The received topic and the extracted utterances can be input to a trained machine learning model. The trained machine learning model generates example utterances for the new intent. The new intent with the generated example utterances can be used as training data for training the chatbot.

Abstract: Chatbot logs can be received. The chatbot logs can be clustered or classified into clusters of potential intents. Trending topics can be detected by analyzing at least social media content. Semantic similarities between the potential intents and the trending topics can be computed. Based on the computed semantic similarities, at least one of the potential intents to at least one of the trending topics can be correlated. Training of the chatbot using the chatbot logs associated with the correlated potential intent can be triggered.

Abstract: A method of automatically creating new intents and assigning examples to those new intents in a dialog system includes generating at least one predicted intent from an intent concept graph by combining two or more nodes of the intent concept graph. A graph embedding is generated of the at least one predicted intent. A sentence embedding is generated from examples from a call log. The graph embedding of the at least one predicted intent is then matched with the examples from the sentence embedding and the at least one predicted intent, with the examples, is connected into a dialog system.

Abstract: A method and system of generating a documentation includes monitoring user actions regarding a model development, wherein the user actions are captured in a source code development environment. Semantic meaning is provided for each user action captured in the source code development environment. A degree of impact of each user action is determined in connection with the model. Actions having a degree of impact in the development of the model that is above a predetermined threshold are identified as impactful actions. An interactive knowledge graph is identified based on the user actions, semantic meaning of each action, and the determined degree of impact of each user action. The interactive knowledge graph is provided to be displayed in a navigable way.

Abstract: Techniques for maintaining a knowledge database system based on user interactions with the database system are described herein. An aspect includes generating a user interface based on a knowledge database comprising a plurality of entities, wherein the user interface comprises a plurality of user interface elements, each user interface element of the plurality of user interface elements corresponding to one or more of the plurality of entities of the knowledge database. Another aspect includes receiving a user interaction corresponding to the user interface. Another aspect includes updating the knowledge database based on the received user interaction, wherein updating the knowledge database based on the received user interaction includes determining a user interface element associated with the received user interaction, mapping the determined user interface element to a first entity of the plurality of entities of the knowledge database, and performing a weight update corresponding to the first entity.

Abstract: A system for initiating web containers and isolating network traffic to web containers is described. The system initiates and uses web containers for associated web-based resources, such as domain addresses owned by a single entity. When a web container for a domain address has not been initiated, the system determines an owner of the domain address and any domain addresses associated with the owner and initiates a web container for the domain address and the associated domains. When a user navigates to the domain addresses associated with the determined owner, the system isolates the network activity to the initiated web container.