Digital Commons Network^™

Rapid Troubleshooting And Management Of A Network Using A Chatbot, N/A

Defensive Publications Series

Managing a large and complex network is currently tedious and error-prone: administrators log in via lengthy authentication processes; look up and access a diversity of network devices, services, tools and other entities with differing communication protocols; issue relatively obscure, device-specific commands; etc. These problems of network management become especially manifest while troubleshooting critical issues under time constraints. This disclosure describes a fast, simple, and unified approach to network management via a chat interface. At the back end of the chat interface is a bot that is in communication with a variety of network entities in their native protocols. The bot …

Embedding Explorer, Anonymous Anonymous

Defensive Publications Series

The present disclosure describes an embedding explorer that allows a user to interactively explore properties of an embedding space and how the embedding space relates to features of entities being embedded. The embedding space is a low-dimensional space onto which the embedding explorer translates high-dimensional vectors to low-dimensional vectors. In particular, the embedding explorer allows the user to load a table of embeddings and feature providers that are required by the user during the exploration. The table includes at least one column with an entity ID and another column with an array of floats representing the embedding for the entity. …

Automated Calculation Of Spacing Constraints In Multi-Layer Pcb Design, N/A

Defensive Publications Series

A multi-layer PCB comprises one or more layers made of conducting shapes sandwiched between dielectric layers. The spacing between conducting shapes on a layer is constrained to optimize signal integrity and minimize cross-talk. Such spacing constraints are typically manually calculated and entered into a design tool that enforces the constraints during the design process. Spacing constraints depend on a variety of parameters, including dielectric thicknesses, type of conducting shapes, signal technology, voltage, waveform type, data rate, etc. Manually calculating and maintaining the spacing constraint set across a multi-dimensional matrix of possible values is onerous and error-prone. Engineers frequently simplify the …

Predicted Application Preload Based On Context, Collin Irwin, Rachel Hausmann, Joe Rothlin

Defensive Publications Series

A computing device (e.g., a mobile phone, camera, tablet computer, etc.) uses an application preload prediction model (e.g., an artificial intelligence model) for preloading an application. The computing device may execute an application that employs a user interface to facilitate human-machine interaction. The computing device may collect contextual user data and use a prediction model (e.g., an application preload prediction module) to analyze the collected contextual data to determine and predict an application to preload in background mode even with the computing device’s screen locked. For example, a user device (e.g. mobile phone) with a touch screen predicts a user’s …

Critical Request For Proposals (Rfps) Requirement Insight Engine (Carrie), Hp Inc

Defensive Publications Series

According to projections from IDC, 80% of worldwide data will be unstructured by 2025. For many large

companies including, its reached the critical mass already. Unstructured data, such as RFPs, creates a

unique challenge for organizations wishing to extract insight and use it to make strategic decisions [1]. We

created a stacked machine learning algorithm based on Natural Language Processing (NLP) to extract

insights from different RFP documents, we named the algorithm CaRRIE for Critical RFPs Requirement

Insight Engine. CaRRIE is able to analyze how similarly-matched an RFP is to different an organization

service offerings thus identifying gaps in it’s …

Adaptative Maintenance, Hp Inc

Defensive Publications Series

In production printers, there are subsystems that require service maintenance to guarantee the full life of

the product as well as the proper performance over time.

If the service maintenances are not followed, there is risk of part failures and unexpected printer downtime.

The only way of reducing the unexpected printer interruptions is by enabling predictive service maintenance

that allow the service organization to schedule the interventions on the printer before the problem occurs.

The new Adaptative maintenance system will allow company to adapt the service maintenance plan and

the on-site interventions based on customer usage to only replace predictively …

Os And Hw Independent Memory Snapshot Tool For Analyzing System State And Reverse Engineering Malware And Bugs, Hp Inc

Defensive Publications Series

A method for obtaining a memory snapshot of a computer while in operation is disclosed. This method

involves using a System Management Interrupt (SMI) to read memory and write it out to a designated

output for further analysis.

Contextual Recipe Recommendations, William Li, Amit Gera

Defensive Publications Series

Online services provide recipe recommendations to users on a variety of endpoint devices. When recipe recommendations are provided based on a user’s long-term preferences, such recommendations may not be best suited to the user’s current context. This disclosure describes techniques to provide context-aware recipe recommendations when users provide permission to determine the context. Context determination can include, e.g., analysis of recent user queries to a virtual assistant or a search engine to identify triggering queries that are associated with food. A machine learning model is used to map queries to user intent. Recipes associated with the determined user intent are …

Automatic Content Filtering In Virtual Assistants For Kids, Yuzhao Ni

Defensive Publications Series

Virtual assistant responses need to be both useful and safe for kids and families. However, this is currently not always the case. For example, virtual assistant responses can sometimes unexpectedly include explicit answers or answers that are otherwise unsuitable for kids. However, restricting searches can prevent the virtual assistant from surfacing useful, family-safe responses. There is no systematic way to filter non-textual media content, e.g., music, video, etc. Per the techniques of this disclosure, a library of content classifiers is provided that filters out various categories of content inappropriate for children, e.g., explicit content, violent content, etc. A query to …

Automated Restaurant Orders Using Digital Assistant, Sriram Bhargav Karnati, Sonali Mangal

Defensive Publications Series

This disclosure describes techniques that enable a user to place orders for food using a digital assistant. An order for restaurant food pickup or delivery is received from a user. The digital assistant obtains additional details of the restaurant, e.g., from the restaurant website with the use of machine learning techniques to extract text from the website or from other information sources. With user permission, a call is made by the digital assistant to the restaurant to confirm availability of requested items and to complete the order. Details such as an amount of payment to be made, estimated time of …

Distributed Location Processing Using Reverse Tracking And An Internet Of Things Platform, Jeff Apcar, Vinay Saini, Ankush Arora

Defensive Publications Series

The concept of location tracking is of great relevance in various enterprise and industrial environments. The techniques presented herein propose location tracking techniques that are independent from centralized servers and that use scalable technology and architectures to provide resiliency, faster location updates, reduced cost, and increases reliability.

Approximating 3d Poly Mesh Virtual Objects With Primitive Shape Colliders, Chloe Snyder, Alper Gungormusler, Andrew Allen, Aleksandr Ayvazov, Dobromir Yordanov

Defensive Publications Series

A seamless immersive experience requires appropriate physical interactions between 3D objects in the scene being viewed. Systems typically handle such interactions by detecting and resolving collisions between 3D objects in the scene. However, collision detection between two high poly meshes of the two colliding 3D objects has the computational order of O(N²) where N is the triangle count for the respective meshes, thus making the computations time-consuming which prohibits real-time use. Current techniques include manual creation of simpler collider shapes such as spheres and boxes to approximate the geometry. This disclosure describes techniques that automatically approximate the poly …

Controlling The Density Of User-Generated Content In Augmented Reality, Austin Mccasland, Elly Nattinger, Seth Raphael

Defensive Publications Series

In a multi-user augmented reality (AR) environment, a user can insert virtual objects that other users can see. If users place objects without constraint, then the AR view can become dense, overwhelming, and hard to understand. Per the techniques of this disclosure, the density of objects in an AR environment is constrained by maintaining a minimum distance between existing and newly-placed objects.

Secure And User-Friendly Display Of Schedule In A Meeting Room, Alessandro Pelosi

Defensive Publications Series

Meeting and conference rooms in offices are equipped with video conferencing equipment that includes a screen that displays meetings scheduled in the room, e.g., on an idle screen display. The displayed schedules can include meeting titles or codes. If a visitor enters the meeting room, such display can potentially reveal information that is confidential to the organization. This disclosure describes techniques to automatically determine whether display of meeting-related information is appropriate. External meetings are identified based on available contextual information, e.g., meeting attendees, agenda/notes, room location, schedules of adjacent rooms, etc. Ahead of external meetings in a room, schedule information …

Output-Dependent Access Control, Brett Krueger

Defensive Publications Series

Access control management techniques can reduce the risk of data exfiltration while making privacy-trivial data insights, such as statistical correlations, more accessible. Controlling access to sensitive or private data without unduly restricting essential activities is challenging. Traditional access control techniques, including whitelists, blacklists, and Access Control Lists (ACLs) limit access to sensitive data even when the query output does not contain sensitive information. This approach uses a combination of differential privacy, policy-based access controls, and dynamic query analysis. Dynamic query analysis determines what access controls apply to the output of query, i.e. what output-dependent access controls are appropriate. When output-dependent …

Media Access Control Security Key Distribution Using Blockchain And Public Key Cryptography, Niranjan M. M, Nagaraj Kenchaiah, Vijay Kothamasu

Defensive Publications Series

Techniques are described herein for sophisticated authentication and encryption methods that do not require manual configuration or a centralized server. These techniques use blockchain and public key cryptography to exchange Media Access Control security (MACsec) keys securely between router links and thereby by avoid manual configuration for MACsec. This simplifies existing MACsec key configuration approaches, which use static security mode with manually-configured security keys and dynamic security mode with keys distributed from a centralized Authentication, Authorization, and Accounting (AAA) server over Extensible Authentication Protocol Transport Layer Security (EAP-TLS).

Signal Quality-Based Buffering Of Media Data Using An Intermediate Device, Jonathan Hurwitz

Defensive Publications Series

This publication describes systems and techniques to buffer media data at an intermediate device based on a channel quality between a media source device and a media presentation device. In some environments with significant user movement, such as the gym or home, media packets may be dropped because a condition of a radio-frequency (RF) channel becomes worse over time. Consequently, media presentation may be interrupted. Users often have, however, multiple devices with different amounts of resources. For example, a user may have a smartphone, a smartwatch, and wireless headphones. The latter two are typically carried on a user’s person, but …

Tur As An Input Device To Communicate With The Printer, Hp Inc

Defensive Publications Series

Large format printers that are capable of printing in roll‐to‐roll mode usually have a Take‐Up Reel

(TUR) that is located in a low position of the printer. For this reason, if the user wants to operate the

TUR, usually it has to bend down to a position from where it is not possible to reach the front panel

of the printer. In these cases, during TUR operation it is not possible to command the printer unless

there are some devices in the TUR module (such as buttons, levers, switches, … that allow the user

to do it. Generally, though, these …

Measuring And Visualizing User Sentiment Changes Over Multiple Channels, Xinjun Zhang, Qihong Shao, Changyong Zhao, Antonio Nucci

Defensive Publications Series

Techniques are provided to simultaneously infer approximately where a speaker is looking and the speaker's emotion during a conversation. Due to privacy concerns, only the speaker's approximate facial features may be estimated. The inferred face may be converted into a cartoon face that retains the main facial features. This may enhance user interaction experience even when a speaker does not turn on video in teleconference.

Providing Targeted Event Recommendations, Henry Scott-Green, Devvret Rishi

Defensive Publications Series

A significant portion of live event tickets go unsold, mainly due to a lack of awareness among prospective attendees. Providing relevant event recommendations to the potential audience for an event is a significant industry challenge. This disclosure describes techniques that issue personalized event or artist recommendations based on factors obtained with user permission. For example, such factors may include user history pertaining to location, entertainment, search, social media, etc.

Mmwave Radar Return For Characterization Of Burned Skin, Sean Korphi, Vincent Mei

Defensive Publications Series

This publication describes techniques and systems directed to classifying skin injury using a radar propagation technique to measure the health of tissue below the surface of the skin. The millimeter-wave (mmWave) frequency of the electromagnetic spectrum can be broadcast from an active sensor(s) (e.g., transmitting antenna(s)) on a computing device towards the skin of a person. The mmWaves penetrate the skin and reflectively scatter differently for various tissues and burn types due to changes in dielectric properties. Sensor(s) (e.g., receiving antenna(s)) on the computing device can sense the scattering of the mmWaves reflected by the skin and collect data for …

Classifying And Separating Messages, Alerts, And Notifications, Yameng Lee Thorp, Hideaki Matsui

Defensive Publications Series

This publication describes techniques and apparatuses that allow a user to classify messages, alerts, and/or notifications based on their significance to the user and allow the user to separate them to be displayed in a first or a second part of a display screen of a smartphone. A machine-learned (ML) model aids the user by classifying and separating messages, alerts, and/or notifications of little or significant interest. By doing so, the user can better-manage their amount of time the user spends in front of the display screen of the smartphone and the user can utilize their smartphone more effectively.

Predicting The Impact Of Maintenance Work On Networks, N/A

Defensive Publications Series

In large or complex distributed systems and networks, it is difficult to predict the impact of a maintenance task. Sometimes, multiple maintenance tasks are carried out concurrently, increasing the chance of outage due to inadvertent interaction between the maintenance tasks. In a bid to minimize the chance of outage, network managers often approach network maintenance in a conservative manner, resulting in a loss in the velocity of maintenance changes. This disclosure describes techniques that automatically vet the impact of network change requests by predicting network topology, modeling demands, simulating the effect of the change request, and reporting the impact back …

Metadata Driven User Interface Generator Based On Device Type, Jasvinder Bhasin, Ganesha Shankara Bhat

Defensive Publications Series

Techniques are described herein that provide an innovation in which each device type within a system may receive its own metadata in order to draw (e.g., display) and populate user interface (UI) elements. Different device types and their corresponding UI elements (e.g., tabs, tables, columns, etc.) are visible from one devices' list page. This enables the system to scale in multiple device categories (e.g., routers, gateways, endpoints, etc.). The system can be extended to include other device categories that a current product may not be managing.

Determining Local Depth From Structured Light Using A Regular Dot Grid, Hart Levy

Defensive Publications Series

This publication describes systems and techniques directed to determining local depth, or depth gradients, of a scene from structured light using a regular dot grid pattern and a single receiving camera. A patterned illuminator with a repeating nature provides a structured light that is used to generate depth information. Because any patch of dots can be expected to repeat at a known horizontal distance, an auto-correlation of the image with itself produces peaks along with these repetitions. By finding the disparity from such a peak to its expected location, we measure the local disparity between pairs of patches in the …

Monitoring And Reporting The Safety Performance Of Different Models And Versions Of Autonomous Vehicles, Jay Johnston, David White, Abdelmoniem Rezk, Magnus Mortensen

Defensive Publications Series

Presented herein are techniques that enable an administrative authority (e.g., city, municipality, etc.) to determine when a vehicle is under autonomous control, as well as to specifically determine the vendor and software version of the autonomous control system controlling the vehicle. If the vehicle has problems navigating the environment, the details of that problem are reported directly to the administrative authority. Any trending problems detected after a software update to the control systems can be proactively detected and reported to the vehicle vendor and upstream aggregation systems.

Automated Interactive Threat Analysis Of It Architectures, Tim Hemel

Defensive Publications Series

Threat modeling or architectural risk analysis (ARA) is a process to find cybersecurity threats in an IT system by analyzing its architecture. Because of the large number of possible threats to consider in such an analysis, it helps to automate it. Automatic application of threat analysis rules gives more consistent results and reduces the dependency on expert knowledge in threat modeling, but asks the threat modeler to annotate the architecture with relevant information. The threat modeler will however still have to know what information to provide. In this disclosure we describe a system that interactively asks the threat modeler to …

Dynamic Onboarding And Management Of Sophisticated Edge Clusters For Industrial Internet Of Things (Iot) Applications, Suresh Sankaran, Umang Tandon

Defensive Publications Series

Certain conventional compute solutions are restricted to resources present in the networking hardware (e.g., routers/switches). However, increasing numbers of edge compute workloads are demanding special requirements with Graphical processing unit (GPU)/tensor processing unit (TPU) support, various in-built Internet of Things (IoT) protocol requirements, high system resources, etc. The techniques presented herein address the market of those specialized edge compute workloads while also extending IoT container orchestration to off-the-shelf devices.

Servo Motor Shutdown Prediction For 3dmjf, Hp Inc

Defensive Publications Series

Goal for Predictive Maintenance for 3D printing system is to contribute to reduce unplanned

downtime and possible failures, helping the customers save money while building trust. Not all

failure modes can be captured in internal testing owing to variety of usage environments and

usage conditions. For instance, the threshold for specific sensor monitoring might not be

applicable with specific usage conditions for customer(s). In this paper, we describe a process

based on machine learning to predict the lifespan of servo motors (Z-Motor) in terms of number

of layers remaining and generate alarms.

Generate Augmented Training Of Physical Objects, Hp Inc

Defensive Publications Series

This idea describes a method to generate training data of physical objects using simulated 3D

models to help generate more robust object classification machine learning models.