Abstract: A bot detector is disclosed. The bot detector can apply one or more subsystems for detecting bots. The subsystems may include one or more of a system for identifying self-identified bots, a system for applying one or more rules to identify bots, or a system for identifying bots based on outlier activity. The system for identifying bots based on outlier activity may include one or more outlier detection models that determine whether a user is an outlier based on features of activity data associated with a website.

Abstract: Methods and systems for performing named entity recognition are disclosed. One method includes using a multi-task approach to fine-tune a neural network to perform named entity recognition. A multi-task objective function can include a combination of a triplet loss and a named entity recognition loss. The triplet loss can include the use of supplementary texts. The method further includes using the fine-tuned neural network to identify one or more named entities in a text. Aspects of the disclosure also include integrating named entity recognition with one or more other natural language processing tasks.

Abstract: Methods and systems for performing multi-task learning of query intent and named entities are provided. One method includes receiving a query comprising query text. The method further includes providing the query text to a neural network model implemented on a computing system, the neural network model having a plurality of layers, wherein at least one layer comprises a plurality of loss functions including a named entity tag learning loss function and an intent classification loss function. The method also includes obtaining, from the neural network model, an identification of a named entity and a query intent derived from the query text. A query response may be formulated based, at least in part, on the named entity and query intent.

Abstract: Systems and methods for automating image annotations are provided, such that a large-scale annotated image collection may be efficiently generated for use in machine learning applications. In some aspects, a mobile device may capture image frames, identifying items appearing in the image frames and detect objects in three-dimensional space across those image frames. Cropped images may be created as associated with each item, which may then be correlated to the detected objects. A unique identifier may then be captured that is associated with the detected object, and labels are automatically applied to the cropped images based on data associated with that unique identifier. In some contexts, images of products carried by a retailer may be captured, and item data may be associated with such images based on that retailer's item taxonomy, for later classification of other/future products.

Abstract: Methods and systems for predicting relevant items to be presented to a user in an online environment are described. The methods and systems described herein generate models based on previous item selections to determine an overall time series model for predicting a relevant time of next item selection as well as items most likely to be selected at that time. Complementary items can be presented to the user alongside the selection of most relevant items.

Abstract: Methods and systems for identifying one or more products in an electronic image are disclosed. The computer-implemented method to electronically recognize a product in an electronic image captured via an electronic mobile device is disclosed. The method includes receiving, by a server, a video stream from a camera of the electronic mobile device, the video stream including a plurality of frames. The server selects at least one of the plurality of frames from the video stream, the at least one of the plurality of frames from the video stream being selected is a captured image. The server segments a plurality of products in the captured image into a plurality of segments. The server performs an image recognition using each of the plurality of segments to identify the product in each of the plurality of segments. One or more recognized products identified in the image recognition is outputted by the server.

Abstract: Systems and methods for automating image annotations are provided, such that a large-scale annotated image collection may be efficiently generated for use in machine learning applications. In some aspects, a mobile device may capture image frames, identifying items appearing in the image frames and detect objects in three-dimensional space across those image frames. Cropped images may be created as associated with each item, which may then be correlated to the detected objects. A unique identifier may then be captured that is associated with the detected object, and labels are automatically applied to the cropped images based on data associated with that unique identifier. In some contexts, images of products carried by a retailer may be captured, and item data may be associated with such images based on that retailer's item taxonomy, for later classification of other/future products.

Abstract: Systems and methods for automating image annotations are provided, such that a large-scale annotated image collection may be efficiently generated for use in machine learning applications. In some aspects, a mobile device may capture image frames, identifying items appearing in the image frames and detect objects in three-dimensional space across those image frames. Cropped images may be created as associated with each item, which may then be correlated to the detected objects. A unique identifier may then be captured that is associated with the detected object, and labels are automatically applied to the cropped images based on data associated with that unique identifier. In some contexts, images of products carried by a retailer may be captured, and item data may be associated with such images based on that retailer's item taxonomy, for later classification of other/future products.

Abstract: Methods and systems for performing named entity recognition are disclosed. One method includes using a multi-task approach to fine-tune a neural network to perform named entity recognition. A multi-task objective function can include a combination of a triplet loss and a named entity recognition loss. The triplet loss can include the use of supplementary texts. The method further includes using the fine-tuned neural network to identify one or more named entities in a text. Aspects of the disclosure also include integrating named entity recognition with one or more other natural language processing tasks.

Abstract: A retail store including a server having a processor and a memory; a communication network; and a database are disclosed. The server includes an electronic product recognizer that receives a video stream including a plurality of frames from a camera of an electronic mobile device. At least one of the plurality of frames is selected as a captured image. A plurality of products in the captured image is segmented into a plurality of segments. Image recognition is performed using each of the plurality of segments to identify the product in each of the plurality of segments. One or more recognized products identified in the image recognition are output. The one or more recognized products identified in the image recognition are configured to be sent to a user device communicable with the server via the communication network, the server configured to cause one or more stickers to be displayed on the user device.

Abstract: Methods and systems for performing multi-task learning of query intent and named entities are provided. One method includes receiving a query comprising query text. The method further includes providing the query text to a neural network model implemented on a computing system, the neural network model having a plurality of layers, wherein at least one layer comprises a plurality of loss functions including a named entity tag learning loss function and an intent classification loss function. The method also includes obtaining, from the neural network model, an identification of a named entity and a query intent derived from the query text. A query response may be formulated based, at least in part, on the named entity and query intent.

Abstract: Systems and methods for automating image annotations are provided, such that a large-scale annotated image collection may be efficiently generated for use in machine learning applications. In some aspects, a mobile device may capture image frames, identifying items appearing in the image frames and detect objects in three-dimensional space across those image frames. Cropped images may be created as associated with each item, which may then be correlated to the detected objects. A unique identifier may then be captured that is associated with the detected object, and labels are automatically applied to the cropped images based on data associated with that unique identifier. In some contexts, images of products carried by a retailer may be captured, and item data may be associated with such images based on that retailer's item taxonomy, for later classification of other/future products.

Abstract: Methods and systems for identifying one or more products in an electronic image are disclosed. The computer-implemented method to electronically recognize a product in an electronic image captured via an electronic mobile device is disclosed. The method includes receiving, by a server, a video stream from a camera of the electronic mobile device, the video stream including a plurality of frames. The server selects at least one of the plurality of frames from the video stream, the at least one of the plurality of frames from the video stream being selected is a captured image. The server segments a plurality of products in the captured image into a plurality of segments. The server performs an image recognition using each of the plurality of segments to identify the product in each of the plurality of segments. One or more recognized products identified in the image recognition is outputted by the server.

Abstract: A retail store including a server having a processor and a memory; a communication network; and a database are disclosed. The server includes an electronic product recognizer that receives a video stream including a plurality of frames from a camera of an electronic mobile device. At least one of the plurality of frames is selected as a captured image. A plurality of products in the captured image is segmented into a plurality of segments. Image recognition is performed using each of the plurality of segments to identify the product in each of the plurality of segments. One or more recognized products identified in the image recognition are output. The one or more recognized products identified in the image recognition are configured to be sent to a user device communicable with the server via the communication network, the server configured to cause one or more stickers to be displayed on the user device.

Abstract: Methods and systems for predicting relevant items to be presented to a user in an online environment are described. The methods and systems described herein generate models based on previous item selections to determine an overall time series model for predicting a relevant time of next item selection as well as items most likely to be selected at that time. Complementary items can be presented to the user alongside the selection of most relevant items.