Abstract: Systems and methods for providing a contextual conversation via a chat agent. The chat agent includes or is in communication with an artificial intelligence (AI) language model (LM). In examples, the chat agent leverages the LM and one or more knowledge bases to obtain prior conversation context and/or other contextual details to assist in generating accurate and relevant chat responses to chat inputs received from the user. In some examples, a user profile is built asynchronously based on descriptive elements extracted from prior conversations. In other examples, granular contextual details of prior conversations relevant to the chat input are identified based on a semantic search. Long-term preferences and/or granular contextual details are obtained and provided to the LM with received chat input to generate a personalized chat response for the user.

Abstract: The present disclosure relates to systems, devices, and methods for a visual exploration of search engine results. The systems, devices, and methods may present, in response to receiving the search engine results for a query to perform a task, a search result webpage with a cascading waterfall layout that includes a plurality of images from a plurality of webpages included in the search results. The search result webpage may also present, adjacent to the plurality of images, structured data corresponding to a selected image from the plurality of images where the structured data includes details from the webpage to complete the task.

Abstract: The present disclosure relates to systems, devices, and methods for a visual exploration of search engine results. The systems, devices, and methods may present, in response to receiving the search engine results for a query to perform a task, a search result webpage with a cascading waterfall layout that includes a plurality of images from a plurality of webpages included in the search results. The search result webpage may also present, adjacent to the plurality of images, structured data corresponding to a selected image from the plurality of images where the structured data includes details from the webpage to complete the task.

Abstract: Techniques for dynamically generating deeplink search results in response to navigational search queries. In an aspect, to address user search queries, a general-purpose search engine is provided in parallel with a dedicated engine for specifically ranking deeplinks. Upon identifying a received query as a navigational query, a parallel query is generated from a common domain and user query, and provided to the dedicated engine. The engine accesses relevant deeplink URL's from a search index, which may be frequently refreshed and updated with the most recent Web contents. Ranking of deeplink URL's may be performed according to an algorithm that processes query-level features and document-level features of URL's to be ranked. In an aspect, the algorithm may be trained from search engine logs and/or Web browser logs, by calculating a Log-based Normalized Discounted Cumulative Gain (LNDCG) designed to quantify relevance of search results to queries based on user click behavior.

Abstract: Techniques for dynamically generating deeplink search results in response to navigational search queries. In an aspect, to address user search queries, a general-purpose search engine is provided in parallel with a dedicated engine for specifically ranking deeplinks. Upon identifying a received query as a navigational query, a parallel query is generated from a common domain and user query, and provided to the dedicated engine. The engine accesses relevant deeplink URL's from a search index, which may be frequently refreshed and updated with the most recent Web contents. Ranking of deeplink URL's may be performed according to an algorithm that processes query-level features and document-level features of URL's to be ranked. In an aspect, the algorithm may be trained from search engine logs and/or Web browser logs, by calculating a Log-based Normalized Discounted Cumulative Gain (LNDCG) designed to quantify relevance of search results to queries based on user click behavior.

Abstract: The relative priorities or weightings of key performance indicators (KPIs) are objectively evaluated for a web service to facilitate determining where efforts should be made in improving the web service. A KPI-taming cost and user engagement variation is determined for each KPI. The KPI-taming cost for a KPI represents a number of engineering man-hours estimated to be required to achieve a unit of KPI improvement for that KPI. The predicted user engagement variation for a KPI represents an improvement in user engagement with the web service estimated to be provided by a certain improvement in that KPI. A KPI-sensitivity is determined for each KPI based on the KPI-taming cost and predicted user engagement variation for each KPI. A weighting may also be determined for each KPI based on the percentage of each KPI's KPI-sensitivity of the sum of KPI-sensitivities for all KPIs.