Abstract: This application relates to apparatus and methods for applying trained machine learning models to determine an item's relevance to a search query. In some examples, a query and data identifying a plurality of items are received. Item attributes for each of the plurality of items are obtained, and features are generated based on the item attributes. Further, a score is generated for each item by applying a trained machine learning model to the corresponding features and the query. Matching attributes are determined for each of the plurality of items based on the corresponding item attributes and the query, and the score of each of the plurality of items is adjusted based on the matching attributes. Further, the ranking data is generated based on the adjusted score of each of the plurality of items. The ranking data may be transmitted to a web server for display of the items.

Abstract: This application relates to apparatus and methods for ranking search results. In some examples, a computing device receives item data identifying items for a search query provided by a user. The computing determines, for each item, a first value based on a relevance of the item to the search query. The computing device also determines user engagement data for the user, where the user engagement data identifies engagements by the user on a webpage. The computing device determines, for each item, a second value based on the user engagement data. In some examples, the computing device determines the first values and second values by executing one or more machine learning models. The computing device further determines a ranking of the plurality of items based on the first values and the second values. The computing may transmit the ranking of the plurality of items, such as to a web server.

Abstract: This application relates to apparatus and methods for applying trained machine learning models to determine an item's relevance to a search query. In some examples, a query and data identifying a plurality of items are received. Item attributes for each of the plurality of items are obtained, and features are generated based on the item attributes. Further, a score is generated for each item by applying a trained machine learning model to the corresponding features and the query. Matching attributes are determined for each of the plurality of items based on the corresponding item attributes and the query, and the score of each of the plurality of items is adjusted based on the matching attributes. Further, the ranking data is generated based on the adjusted score of each of the plurality of items. The ranking data may be transmitted to a web server for display of the items.

Abstract: This application relates to apparatus and methods for ranking search results. In some examples, a computing device receives item data identifying items for a search query provided by a user. The computing determines, for each item, a first value based on a relevance of the item to the search query. The computing device also determines user engagement data for the user, where the user engagement data identifies engagements by the user on a webpage. The computing device determines, for each item, a second value based on the user engagement data. In some examples, the computing device determines the first values and second values by executing one or more machine learning models. The computing device further determines a ranking of the plurality of items based on the first values and the second values. The computing may transmit the ranking of the plurality of items, such as to a web server.

Abstract: An example method comprises receiving a first data stream regarding performance of a monitored system at a first time, determining a plurality of distributions from the first data stream, identifying at least one state for each different distribution of the plurality of distributions to identify a plurality of states, classifying each of the plurality of states into classifications, identifying at least one of the plurality of states as being a problematic state, for each state recognizing one or more transitions from or to other states of the plurality of states, receiving a second data stream of the monitored system at a second time, identifying a precursor state of the plurality of states indicating at least a potential future transition to the problematic state, and generating a warning before the monitored system enters the problematic state.

Abstract: An example method comprises receiving a first data stream regarding performance of a monitored system at a first time, determining a plurality of distributions from the first data stream using a density function of a plurality of bins for the data, identifying at least one state for each different distribution of the plurality of distributions to identify a plurality of states, classifying each of the plurality of states into classifications, identifying at least one of the plurality of states as being a problematic state using a first log likelihood ratio, for each state recognizing one or more transitions from or to other states of the plurality of states, receiving a second data stream of the monitored system at a second time, identifying a precursor state indicating at least a potential future transition to the-problematic state, and generating a warning before the monitored system enters the problematic state.

Abstract: A method to reduce memory requirements for a packet loss concealment algorithm in the event of packet loss in a receiver of pulse code modulated voice signals. A voice playout unit in the receiver shares its nominal delay buffer with a history buffer of a packet loss concealment algorithm up to a maximum limit described in a standard. This reduces or eliminates need to allocate memory for the history buffer. A history buffer can also be extended to retain an original portion of voice signal packets received prior to a packet loss as well as generated voice signals as they are generated. A scratch buffer is used as a working buffer and replaces the function of a pitch buffer.

Abstract: A method to reduce memory requirements for a packet loss concealment algorithm in the event of packet loss in a receiver of pulse code modulated voice signals. A voice playout unit in the receiver shares its nominal delay buffer with a history buffer of a packet loss concealment algorithm up to a maximum limit described in a standard. This reduces or eliminates need to allocate memory for the history buffer. A history buffer can also be extended to retain an original portion of voice signal packets received prior to a packet loss as well as generated voice signals as they are generated. A scratch buffer is used as a working buffer and replaces the function of a pitch buffer.

Abstract: A method to reduce memory requirements for a packet loss concealment algorithm in the event of packet loss in a receiver of pulse code modulated voice signals. A voice playout unit in the receiver shares its nominal delay buffer with a history buffer of a packet loss concealment algorithm up to a maximum limit described in a standard. This reduces or eliminates need to allocate memory for the history buffer. A history buffer can also be extended to retain an original portion of voice signal packets received prior to a packet loss as well as generated voice signals as they are generated. A scratch buffer is used as a working buffer and replaces the function of a pitch buffer.

Abstract: A method to reduce memory requirements for a packet loss concealment algorithm in the event of packet loss in a receiver of pulse code modulated voice signals. A voice playout unit in the receiver shares its nominal delay buffer with a history buffer of a packet loss concealment algorithm up to a maximum limit described in a standard. This reduces or eliminates need to allocate memory for the history buffer. A history buffer can also be extended to retain an original portion of voice signal packets received prior to a packet loss as well as generated voice signals as they are generated. A scratch buffer is used as a working buffer and replaces the function of a pitch buffer.

Abstract: A fax-aware telephone for handling facsimile transmissions over a computer network, such as the Internet, that includes a facsimile processing unit for handling an incoming or outgoing facsimile call, a voice processing unit for handling an incoming or outgoing voice call, a memory unit for storing an incoming facsimile transmission, and a file transfer protocol client for transferring a facsimile transmission to the computer network, an external storage device, or to a printer driver. The fax-aware telephone handles a facsimile call and a voice call together over a single data line from a computer network.