Abstract: Some embodiments provide a method for an agent executing on a Kubernetes node in a cluster. The method instructs a forwarding element that also executes on the node to process a flow tracing packet. From the forwarding element, the method receives a message indicating a set of flow entries matched by the flow tracing packet as the forwarding element processes the flow tracing packet. For each flow entry of at least a subset of the flow entries matched by the flow tracing packet, the method generates mapping data that maps elements of the flow entry to Kubernetes concepts implemented in the cluster. The method reports data regarding the set of flow entries along with the generated mapping data.

Abstract: The invention provides a method and system for training a machine learning-based patent search or novelty evaluation system. The method comprises providing a plurality of patent documents each having a computer-identifiable claim block and specification block, the specification block including at least part of the description of the patent document. The method also comprises providing a machine learning model and training the machine learning model using a training data set comprising data from said patent documents for forming a trained machine learning model. According to the invention, the training comprises using pairs of claim blocks and specification blocks originating from the same patent document as training cases of said training data set.

Abstract: A system and method for technology analysis utilizing high-performance, scalable, multitenant, dynamically specifiable, knowledge graph information storage and utilization. The system uses an in-memory associative array for high-performance graph storage and access, with a non-volatile distributed database for scalable backup storage, a scalable, distributed graph service for graph creation, an indexing search engine to increase searching performance, and a graph crawler for graph traversal. One or more of these components may be in the form of a cloud-based service, and in some embodiments the cloud-based services may be containerized to allow for multitenant co-existence with no possibility of data leakage or cross-over. The system uses a cyber-physical graph to represent an enterprise's cyber-physical system and can provide graph analysis, graph security, and graph fusion related tasks to identify potential operational risks.

Abstract: Methods, systems, and media for identifying abusive user accounts based on playlists are provided. In accordance with some embodiments of the disclosed subject matter, a method for identifying abusive content is provided, the method comprising: determining at least one feature associated with a playlist created by a user-generated channel; calculating a playlist score associated with the playlist based on a playlist classifier, wherein the playlist classifier comprises a function that maps the at least one feature to the playlist score; calculating a channel score associated with the user-generated channel based at least on the calculated playlist score; determining that one or more content items associated with the user-generated channel is to be demoted based on the calculated channel score, wherein the one or more content items comprises the playlist; and causing the one or more content items to be demoted.

Abstract: An apparatus and method for using a data mesh to structure unstructured data is provided.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for associating different watermarks detected in media. An example method disclosed herein includes determining whether a first watermark detected in a media signal is represented in a watermark data structure and associating the first watermark with a first media presentation record associated with a second watermark in response to the first watermark being associated in the watermark data structure with the second watermark. The example method further includes transmitting monitoring data including the first media presentation record to an audience measurement entity.

Abstract: Embodiments are directed to a snap point segmentation that defines the locations of selection snap points for a selection of video segments. Candidate snap points are determined from boundaries of feature ranges of the video indicating when instances of detected features are present in the video. In some embodiments, candidate snap point separations are penalized for being separated by less than a minimum duration corresponding to a minimum pixel separation between consecutive snap points on a video timeline. The snap point segmentation is computed by solving a shortest path problem through a graph that models different snap point locations and separations. When a user clicks or taps on the video timeline and drags, a selection snaps to the snap points defined by the snap point segmentation. In some embodiments, the snap points are displayed during a drag operation and disappear when the drag operation is released.

Abstract: The present disclosure is directed to managing and accessing experiment data. A data processing system may generate a system identifier for each instance that leads to a generation of experiment data. A first segment may refer to a source entity, and may be used to differentiate identifiers generated by different source entities. This implementation may aim to provide an explicit representation of the data generation process while providing a glimpse on select metadata attributes, such as an individual species, sample class, and experimental technique used to generate the dataset. These constraints may prevent duplicates and may enable system-side associations among subjects, samples, aliquots, and experiments.

Abstract: A system and method may classify a plurality of interactions, by: obtaining a plurality of voiceprints of the plurality of interactions, wherein each voiceprint of the plurality of voiceprints represents a speaker participating in an interaction of the plurality of interactions; calculating, for each interaction, a plurality of scores, wherein each score of the plurality of scores is indicative of a similarity between the voiceprint of the interaction and one voiceprint of a set of benchmark voiceprints; calculating, for each interaction, statistics of the scores; and determining that a plurality of interactions pertain to a single cluster of interactions based on statistics of the scores of the interactions in the cluster.

Abstract: Systems, methods, and devices for obfuscation of browser fingerprint data are disclosed. In one embodiment, in an information processing apparatus comprising at least one computer processor, a method may include: (1) receiving an electronic communication from an electronic device to a destination; (2) identifying a raw device fingerprint value in the electronic communication from an electronic device; (3) selecting at least one replacement value for the raw device fingerprint value; (4) replacing the raw device fingerprint value with the replacement value in the electronic communication; and (5) sending the electronic communication with the replacement value to the destination.

Abstract: The present invention relates to a support vehicle (20) for performing support operations in an automated storage and retrieval system (1). It comprises a vehicle body (21). It comprises a drive system (40) comprising wheels (42, 44) provided in a lower part of the vehicle body (21), where the drive system (40) being configured to drive the support vehicle (20) along a track system (108) of the automated storage and retrieval system (1). It comprises a connection system (30) provided on a first side (20a) of the support vehicle (20). The connection system (30) comprises a connector member (31) protruding through a aperture (22) of the vehicle body (21). The connection system (30) comprises an actuator (34) for moving the connector member (31) in the aperture (22) in relation to the vehicle body (21). The present invention also relates to an automated storage and retrieval system (1).

Abstract: The present disclosure generally relates to drink reminder, and more particularly, to a wearable reminder for drinking in a bracelet or a finger ring form factor. One of issues that cause illness in elderly or persons with special illness or ordinary persons due to improper habit, is frequently they forget to drink enough amounts of liquid and this inadequate drink lead to dehydration. Using special drinkware which comprises bulky circuits and batteries is not convenient for users and people like to use their own personal drinkware, mug, cup or glass. This disclosure proposes a wearable reminder for drinking and can be used with any kind of drinkwares and only need to attach an active, passive or semi-active RFID tag to the drinkware. The user wears a wearable like bracelet or a finger ring and wearable reminds him/her based on predefined time schedule or condition.

Abstract: In accordance with an embodiment, described herein is a system and method for providing sandboxing support in a multidimensional database environment. A plurality of sandboxes and a baseline data can be provided in a multidimensional database server, with each sandbox created for a particular “what-if” analysis, and configured to store one or more changes to the baseline data. When a request is received for a report on a particular “what-if” analysis from a client/user, the multidimensional database server can split the request into a first query and the second query, with the first query directed to a sandbox associated with the client/user, and the second query directed to the baseline data. The multidimensional database server can merge results from the first query and the second query, and perform aggregations and calculations on the merged data, before sending the merged data to a client.

Abstract: Systems and methods for generating media program keywords based on a video-hosting website are disclosed herein. Control circuitry identifies, on the video-hosting website, video content items that include at least a portion of a media program. The media program has a media program identifier and the video content items have respective titles, each including one or more terms. The control circuitry identifies a term included in more than one of the titles and identifies a group of the video content items that have the term included in their title. Based on the video-hosting website, the control circuitry determines a cumulative number of rankings of the video content items within the group and generates a relevance score for the term based on the cumulative number of rankings. The control circuitry stores the term and the relevance score in a keyword database in association with the media program identifier.

Abstract: A media application determines, based on pixels of images or videos from a collection of media items, clusters of media items such that the media items in each cluster have a visual similarity, wherein the collection of media items is associated with a user account. The media application selects a subset of the clusters of media from corresponding clusters of media items based on the media items in each cluster having a visual similarity within a range of threshold similarity values. The media application causes a user interface to be displayed that includes the subset of the clusters of media.

Abstract: Systems and methods for authentication and authorization using physical computing cards include receiving from the physical computing card an authorization request, where the request includes a card identifier and an entity location. Authorized users associated with the card identifier are determined. Identifications of users with pending transactions at the entity location are obtained. Verification is performed by determining that there is a match between an identification of a user to an identification of one of the authorized users. The user is then authorized to access resources associated with the physical computing card.

Abstract: Provided are a system and method for distributing the digitization of a payment credential to an entity that does not have access to issuer master keys. In one example, the method includes receiving a personalization script for a payment credential, the personalization script generated by an issuer of the payment credential, generating tokenized payment information for the payment credential based on the received personalization script, wherein the tokenized payment information is generated by the digitization entity without having access to issuer master keys (IMKs) of the issuer of the payment credential, and transmitting the tokenized payment information for the payment credential to a digital wallet executing on a user device.

Abstract: A system and method for launching gameplay from streaming content is provided. Information regarding media content may be stormed in memory. Such media content may have at least one or more trigger points and each trigger point may be associated with a set of game data specific to a gameplay scene within an interactive title. The media content may be streamed to a user device over a communication network. A selection of one of the at least one or more trigger points may be received over the communication network from the user device. Identifying the set of game data associated with the selected trigger point may be identified and the interactive title for gameplay may be launched by the user device based on the identified set of game data associated with the selected trigger point.

Abstract: The invention relates to a method for remotely handling a conspicuity pattern or error report of a means of transport (1a-1d) comprising the steps of: receiving the conspicuity pattern of error report (ER) from the means of transport (1a-1d) on a back-end server (3); automatically determining an error pattern (F1-F6) stored on the back-end server (3) as a function of the error report (ER), in response thereto automatically determining a configuration requirement as a function of the stored knowledge; automatically generating and sending a configuration request representing a union set of the respective configuration requirements to the means of transport (1a-1d); receiving (500) a configuration specification in response to the configuration request on the back-end server (3); and automatically determining an action, like e.g. a troubleshooting operation or an information based on the configuration specification by the back-end server (3).

Abstract: Computer-implemented techniques for unsupervised event extraction are provided. In one instance, a computer implemented method can include parsing, by a system operatively coupled to a processor, unstructured text comprising event information to identify candidate event components. The computer implemented method can further include employing, by the system, one or more unsupervised machine learning techniques to generate structured event information defining events represented in the unstructured text based on the candidate event components.

Abstract: The present invention extends configuring development and operations pipelines using drag and drop techniques. Through user interface gestures, users can form and manage pipelines that span any combination of: public cloud resources, private cloud resources, user on-premise resources, etc., in accordance with appropriate (cloud and/or on-premise) profile information. The user interface facilitates pipeline (re)configuration as appropriate to address alterations to workflows, upgrades to DevOps tools, removal of functionality from a workflow, etc. The user interface enables customers to build no-code pipelines for various use cases in a drag and drop manner. Users can integrate collaboration tools, notifications, and approval gates offering thresholds at each and every step. In addition, the pipeline framework captures logs and provides a summary via livestream and also upon completion of each pipeline activity and after each pipeline.

Abstract: A system for enforcing projection constraints on data values stored in specified variables of a shared dataset of a cloud data platform. A request is received from a first account of the cloud data platform that identifies a first operation to be performed on the shared dataset. A first set of data, including data accessed from a first variable, is accessed from the shared dataset to use in performing the first operation. A projection constraint policy attached to the first variable of the shared dataset is determined, and the projection constraint policy is further determined to be enforced based on the request. Based on the first set of data and the first operation, an output to the first request is generated.

Abstract: Systems, devices, apparatuses, components, methods, and techniques for saving media content to a context for later playback are provided. An example media-playback device for identifying and playing media content for a user traveling in a vehicle includes a context detecting device, a context-driven playback engine, and a media playback engine. Contexts are established by parameters that can be detected by a media-playback device. Contexts are situations that are defined by one or more locations, times, events, activities, people, and devices. Media content is saved to the contexts for later playback. The contexts are detected by the context detecting device, the associated media content is identified by the context-driven playback engine, and the media content is automatically played through the media playback engine, without additional input required by the user.

Abstract: A recommendation method includes determining one or more aspects of a first item based on at least one descriptive text of the first item. The recommendation method also includes updating a knowledge graph containing nodes that represent multiple items, multiple users, and multiple aspects. Updating the knowledge graph includes linking one or more nodes representing the one or more aspects of the first item to a node representing the first item with one or more first edges. Each of the one or more first edges identifies weights associated with (i) user sentiment about the associated aspect of the first item and (ii) an importance of the associated aspect to the first item. In addition, the recommendation method includes recommending a second item for a user with an explanation based on at least one aspect linked to the second item in the knowledge graph.

Abstract: A query syntax analysis and postprocessing system and method receives a query request specifying values of a data type directed to a database without native support for the queried data type. A query engine or process for receiving a query request defined by a query syntax traverses the query request for a specification of an unsupported value or expression. In a large, distributed database environment, the query request may implicate multiple physical data stores, each having specific formats and recognized syntax. Values of unsupported or non-native types, or expressions that evaluate to an unsupported or non-native type, are identified and replaced with an expression that is recognized by the legacy database.

Abstract: Techniques are disclosed for objectively assessing the driving performance of rideshare drivers and/or vehicles. The techniques include collecting data from various sources to aggregate trip data for several drivers and passengers over several ridesharing trips. This data may include telematics data collected from several passengers and/or drivers during each of their respective ridesharing trips, which indicates various aspects associated with operation of the vehicle. Each driver may then be correlated to his own set of trip data, such that a driving assessment may be made for each individual driver. When a new user subsequently requests a ride via a ridesharing provider, this driver assessment may be made available to the user, thereby providing the user with an objective assessment of the driver prior to the start of the scheduled ride.

Abstract: Disclosed are systems and methods that determine a likelihood that an item page represents a risky item. For example, a machine learning model may be trained to determine a probability score that an item page corresponds to a topic of a plurality of topics based on item textual descriptions determined from the item page and search queries corresponding to the item page. Still further, one or more of a topic risk score, item risk score, cumulative item risk score, and/or a total item risk score may be determined. Each of the risk scores may be indicative of a likelihood that the item page represents a risky item.

Abstract: Systems and methods for determining the location of advertisements in multimedia assets are disclosed. A method includes obtaining an audio signature corresponding to a time period of a multimedia asset, identifying a match between the obtained audio signature and one or more stored audio signatures, comparing programming data of the multimedia assets of the obtained audio signature and the matching audio signatures, and determining whether the time period of the multimedia asset contains an advertisement based on the comparison of the programming data of the multimedia assets of the obtained audio signature and the one or more matching audio signatures. Another method includes identifying matches between a plurality of obtained audio signatures and a plurality of stored audio signatures, and determining whether consecutive time periods of the multimedia asset contain an advertisement based on a number of consecutive matching audio signatures of the plurality of stored audio signatures.

Abstract: A question generation device includes: generating means which uses a query and a relevant document including an answer to the query as input and, using a machine learning model having been learned in advance, generates a revised query in which a potentially defective portion of the query is supplemented with a word included in a prescribed lexical set.

Abstract: Systems and techniques that facilitate keyword-object taxonomy generation and utilization are provided. In various embodiments, a system can comprise a receiver component that can access an input object class. In various aspects, the system can comprise a taxonomy component that can output one or more keyword combinations that are non-redundant and relevant to the input object class, based on querying a keyword-object taxonomy. In various instances, the receiver component can access (and/or be provided with an electronic link to) a set of recorded keyword combinations and a set of recorded object classes respectively corresponding to the set of keyword combinations. In various cases, the taxonomy component can generate the keyword-object taxonomy based on the set of recorded keyword combinations and the set of recorded object classes.

Abstract: A method of filtering a URL against a blacklist includes receiving at least a portion of a Uniform Resource Locator (URL), and determining which of a plurality of XOR filters is applicable to the received at least a portion of a URL, where each of the plurality of XOR filters represents a different portion of a URL blacklist. At least a portion of a URL is forwarded to the applicable one of the plurality of XOR filters, and the at least a portion of the URL is processed in the applicable one of the plurality of XOR filters to produce an output indicating whether the URL is likely on the blacklist.

Abstract: The present disclosure is directed towards benchmarking online activity via encoded links generated by an online activity benchmarking system. The method may include identifying by a server of an online activity benchmarking system, for an information resource, a plurality of encoded links encoded by the server of the online activity benchmarking system and linked to the information resource, receiving by the server, via the identified plurality of encoded links, a plurality of requests to access the information resource, identifying by the server, for each request of the plurality of requests to access the information resource, one or more attributes corresponding to the request, categorizing by the server, the plurality of requests to access the information resource based on the one or more identified attributes, and providing by the server, for presentation, an output indicating statistics corresponding to the categorized plurality of requests based on the one or more identified attributes.

Abstract: Systems and methods herein describe privacy preserving multi-touch attribution. The described systems access a plurality of impression events and a plurality of conversion events, and for each impression event and each conversion event, wherein each impression event and each conversion event are associated with user identifiers, the described systems generates a hashed user identifier based on the associated user identifier, initiates a key agreement protocol comprising a key, generates an encrypted identifier by encrypting the hashed user identifier with the key, and stores the encrypted identifier.

Abstract: A method for operating an access control device having a processor, a memory, and an internet connection comprises receiving a credential from a user; checking, by comparing the credential to matching credential information that would indicate the user is authorized to enter stored in the memory of the access control device, if no matching credential information that indicates the user is authorized is found in the memory of the access control device, then checking, at a remote cloud server, a list of buffered credential information for the matching credential information that indicates the user is authorized, and if the matching credential information that indicates the user is authorized is found, sending an unlock signal from the remote cloud server to the access control device.

Abstract: Systems and methods for tracking data lineage and record lifecycle using a distributed ledger are disclosed. In one embodiment, a method for tracking record lifecycle events may include: (1) creating, by a record lifecycle tool, a recordable artifact for a record lifecycle event in a record lifecycle, the recordable artifact comprising data for the record lifecycle event; (2) generating, by the record lifecycle tool, a hash of the data; (3) creating, by the record lifecycle tool, record lifecycle event metadata for the recordable artifact; (4) signing, by the record lifecycle tool, the record lifecycle event metadata; (5) writing the record lifecycle event metadata to supply chain metadata storage at a storage location, wherein the supply chain metadata storage may be cryptographically verifiable and immutable; and (6) writing the hash and an identifier for the storage location in the supply chain metadata store to a present state database.

Abstract: A system, processes, and a computer-readable storage medium are provided method includes receiving a search query including one or more of a textual phrase, a document, or data including metadata from a user. For example, a processor may execute a theme-generating machine learning algorithm to generate a theme of the search query. Profiles of other users are identified as similar to a profile of the user may be located. A search history of each of the located other users may be located in the collected search history. Content in the searchable content that corresponds to the generated theme of the search query and according to a result of the evaluation of the located search history of each of the other users may be obtained. The obtained content may be scored based on a scoring algorithm and output a set of scored search results to the input/output device.

Abstract: Implementations of the present disclosure include actions of receiving a search query, identifying potential search results responsive to the search query, the potential search results corresponding to digital content stored in one or more computer-readable storage media, determining that the potential search results include user-generated content that is generated using one or more computer-implemented social services, receiving data associated with the user-generated content, the data including one or more scores, determining, based on the one or more scores, that the user-generated content is to be provided as a search result, generating search results, the search results including web-based search results and at least a portion of the user-generated content, and transmitting the search results to a client computing device for display to the searching user.

Abstract: Devices, systems and methods are provided to securely and wirelessly connect a medical device to a controller or smartphone with medical device control app, and to wirelessly connect the controller or smartphone with medical device control app to other devices providing additional integrated disease management (IDM) functions. At least two different wireless communication protocols are used depending on level of security needed, with greater security needed for exchange of medical device control operation commands and related data between the medical device and its controller or smartphone with medical device control app to avoid nefarious or unintended changes in medical device control. Exchanges of historical data or notifications relating to the medical device and requiring less security can be transmitted to the controller and/or the smartphone or other IDM devices using a different and more commonly used wireless communication protocol that may afford less security.

Abstract: Methods, systems, and computer-readable media are provided for collective reconciliation. In some implementations, a query is received, wherein the query is associated at least in part with a type of entity. One or more search results are caused to be generated based at least in part on the query. Previously generated data is retrieved associated with at least one search result of the one or more of search results, the data comprising one or more entity references in the at least one search result corresponding to the type of entity. The one or more entity references are ranked, and an entity result is selected from the one or more entity references based at least in part on the ranking. An answer to the query is provided based at least in part on the entity result.

Abstract: Aspects and features of the present disclosure can reconcile node sharing within a production rule network that is fully coded in a compiled language. As an example, the shared, stateless class can represent constraints shared by the alpha node of a rete network. Code can be post processed to create a shared stateless class defined in memory. When the rule engine is executed and the rule network is produced, the shared stateless class can be referenced to evaluate a constraint shared by a node of the production rule network, reducing the number of classes stored in memory. Garbage collection can be used within the shared stateless class, deleting objects from memory structure when no longer used, further improving storage efficiency.

Abstract: One or more techniques and/or computing devices are provided for cross-platform replication. For example, a replication relationship may be established between a first storage endpoint and a second storage endpoint, where at least one of the storage endpoints, such as the first storage endpoint, lacks or has incompatible functionality to perform and manage replication because the storage endpoints have different storage platforms that store data differently, use different control operations and interfaces, etc. Accordingly, replication destination workflow, replication source workflow, and/or a proxy representing the first storage endpoint may be implemented at the second storage endpoint comprising the replication functionality.

Abstract: Systems, methods, and non-transitory computer readable media with instructions for methods here may be used to create and render interactive charts using large amounts of data, by efficiently dividing up the storage and processing of the data between a networked side and a local application side. In certain example embodiments, these two sides work in conjunction with one another, alternatively or additionally, in certain example embodiments, they may act independently and/or alone. The resulting charts may then be used by being embedded in external web pages.

Abstract: One or more techniques and/or systems are provided for implementing a pipeline used to generate, train, test, and implement a document scoring model for assigning document scores to documents. Features from various sources are combined to create a joined page level feature set, a joined domain level feature set, and a host level feature set. Numerical features and content features are extracted from ground truth documents and random documents. The numerical features are joined with the joined feature sets to create a set of joined features. The document scoring model is trained using the set of joined features and a training technique. A document is scored with a document score using the document scoring model based upon the content features and the set of joined features with document scores obtained during training.

Abstract: Various aspects described herein relate to presenting drug dispensing information. Data related to a plurality of dispensing events initiated by one or more employees, of an electronic drug dispensing system can be received. A set of dispensing events of the plurality of dispensing events can be determined as constituting possible misappropriation of drugs by the one or more employees. An alert related to the set of dispensing events can be provided based on determining that the set of dispensing events constitute possible misappropriation of drugs.

Abstract: A server receives from a browser executing on a client device an HTTP request. The server transmits a response to the HTTP request to the browser. The response includes code that when executed by the browser, executes a non-HTTP layer 7 protocol client that communicates with a non-HTTP layer 7 protocol service at an external network. The server receives, from the non-HTTP layer 7 protocol client executing in the browser, data related to the non-HTTP layer 7 protocol service. The server proxies the data related to the non-HTTP layer 7 protocol service over a layer 4 tunnel that is interfaced with the non-HTTP layer 7 protocol service. The server logs event data received from the non-HTTP layer 7 protocol client executing in the browser.

Abstract: A method for processing offline cryptocurrency transfers includes: receiving, by a receiver of a computing device, a first transfer message, wherein the first transfer message is cryptographically signed using a first private key of a first key pair; validating, by a processor of the computing device, the cryptographic signature of the first transfer message using a first public key of the first key pair; storing, in a memory of the computing device, the validated first transfer message; receiving, by an input device of the computing device, a transfer instruction, the transfer instruction including at least a communication address; and electronically transmitting, by a transmitter of the computing device, the validated first transfer message to an external device based on at least the communication address.

Abstract: Systems and methods that are adapted for automatic curation of query responses are disclosed herein. An example method includes obtaining user action metrics corresponding to responses provided in reply to a query for a target resource, the query having a search term, determining a portion of the responses having user action metrics with statistical significance, generating a list of curated responses based on the portion of the responses, and providing the curated responses in reply queries having the search term.

Abstract: A system and method for extending compression-aware aggregation logic to column partitioned database sources when an SQL query involves simple or complex aggregate expressions. The logic can be applied when there are multiple fields specified in a Group By clause, when a Group By clause includes an expression involving multiple columns from a column partitioned table, or when there is no Group By clause in the query. This logic extends the benefits of push-down aggregation to complex aggregate queries to build partially aggregated rows that can be directly added into an intermediate cache. For cases where the fields within aggregate expressions are themselves compressed, the aggregation techniques leverage the compression information of the aggregate fields. This aggregation mechanism can be applicable to compression techniques including run-length encoding (RLE), value list compression (VLC) and Presence, Delta on Mean (PDM) on columnar source tables such as Column Partitioned (CP) or Parquet tables.

Abstract: A method of processing a query to a database from a query source is provided, comprising: receiving the query, the query in a first format supported by the query source; inputting the query into a first neural network; outputting, by the first neural network, the query in a second format, wherein the second format is a format supported by the database; receiving, from the database, a response to the query, the response in the second format; inputting the response to the query into a second neural network; outputting, by the second neural network, the response to the query in the first format; wherein each neural network is trained by inputting a first plurality of pairs of semi-structured data, each pair of semi-structured data comprising a sample query or response in the first format and the sample query or response in the second format.

Abstract: System and methods include obtaining Wi-Fi network data, over the Internet, associated with a plurality of Wi-Fi networks each Wi-Fi network having one or more access points and each Wi-Fi network being associated with a customer of one or more service providers and obtaining customer data for each customer associated with the plurality of Wi-Fi networks, the customer data including call-ins made by customers; aggregating and filtering the data; analyzing the aggregated and filtered data including correlating the call-ins made by customers to the Wi-Fi network data; predicting customer call-ins based on correlations made between the call-ins made by customers and the Wi-Fi network data; and initiating a customer outreach workflow prior to a predicted customer call-in.