Walmart Apollo, LLC

Wal-mart Stores, Inc.

Varun Srivastava

Yan Zheng

Yiye Ruan

Shenghua Wan

Prijith Chandran

Abhishek Gattani

Kannan Govindan

Diego Alvaro Goyret

Parikshit Sondhi

Deepak Arora

Prashant Gupta

Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform: for each respective record in a plurality of records, assigning the respective record: (1) to a first database cluster on a first database server; or (2) to a second database cluster on a second database server; receiving a search request from a requester; executing the search request in the first database cluster to retrieve a first set of results; when the first set of results is greater than a threshold number of results, presenting the first set of results to the requester; and when the first set of results is less than the minimum number of results: executing the search request in the second database cluster to create a second set of results; and presenting the second set of results to the requester, wherein: the threshold number of results is configured such that a probability that the first set of results is greater than the minimum number of results is at least fifty percent. Other embodiments are disclosed herein.

System and method for prioritized product index searching

Various embodiments can comprise a system, which can comprise one or more processing modules; and/or one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of: receiving a request from a requesting party to produce a feature vector for a record, the feature vector being used to aid in placing the record in one of a first database cluster H or a second database cluster L; retrieving a first access history for the record at a first time period, the first access history indicating a number of times the record was accessed within the first time period that is a predetermined time period away from a current day; repeating retrieving one or more second access histories for the record for one or more additional time periods, each second access history of the one or more second access histories indicating a number of times the record was accessed within the one or more additional time periods, wherein each of the one or more additional time periods are different predetermined time periods away from the current day; generating the feature vector based on the first access history and the one or more second access histories of the record; and sending the feature vector to the requesting party. Other embodiments are disclosed herein.

System and method for capturing seasonality and newness in database searches

A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform tracking respective duration data for each respective item of items in a catalog based on purchase histories; measuring a reorder rate for the each respective item within one or more first periods of time; generating a Weibull distribution for the each respective item; training a machine learning model based on previous orders by the users; generating, using the machine learning model, as trained, a ranked list of one or more first items for a user of the users, a respective predicted quantity for each of the one or more first items, and an average basket size for the user; receiving a request for recommended items from the user using a user interface; and sending the ranked list of the one or more first items to be displayed on the user interface. Other embodiments are disclosed.

Smart basket recommendations for online selections

A system and method for predicting search term popularity is disclosed herein. A database system may comprise a first database cluster H and a second database cluster L. A machine learning algorithm is trained to create a predictive model. Thereafter, for each record in a database system, the predictive model is used to calculate a probability of the record being accessed. If the calculated probability of the record being accessed is greater than a threshold value, then the record in the first database cluster H; otherwise, the record is placed in the second database cluster L. Training the machine learning algorithm comprises inputting a training feature vector associated with the record into the machine learning algorithm, inputting a cost vector into the machine learning algorithm, and iteratively operating the machine learning algorithm on each record in the set of records to create a predictive model. Other embodiments are also disclosed herein.

System and method for calculating search term probability

Various embodiments include a system that can comprise one or more processing modules; and one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of: determining a popularity factor for each record in a plurality of records using historical data that can comprise previous search data for each record in the plurality of records; assigning each record in the plurality of records to a first database cluster H on a first database server when the popularity factor for the record in the plurality of records is greater than a threshold value or to a second database cluster L on a second database server when the popularity factor for the record in the plurality of records is less than the threshold value, wherein the first database cluster H has a greater processing power than the second database cluster L; receiving a search request from a requester; executing the search in the first database cluster H to retrieve a first set of results; when the first set of results is greater than a minimum number of results, presenting the first set of results to the requester; and when the first set of results is less than a minimum number of results: executing the search in the second database cluster L to retrieve a second set of results; and presenting the first set of results and the second set of results to the requester. Other embodiments are also disclosed herein.

System and method for prioritized index searching

Various embodiments can include a system that can comprise: one or more processing modules and one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules. The computing instructions can be configured to perform acts of: accessing a query log, wherein the query log can comprise click-through information and item-view information for a record in a set of distinct records in a database system, wherein the database system can comprise a first database cluster H on a first database server and a second database cluster L on a second database server, wherein: the first database server can have a greater processing power than the second database server; the click-through information can comprise information regarding how many times the record was clicked by any user; and the item-view information can comprise information regarding how many times the record was transmitted to be displayed on any display device of any user; calculating a total click-through for the record by determining how many times the record was clicked by any user during a time period; calculating a total item-view for the record by determining how many times the record was transmitted to be displayed on any display device of any user during the time period; storing the record in either the first database cluster H or the second database cluster L based on the total click-through for the record and the total item-view for the record; receiving a search request from a requester; and transmitting a search result to be displayed on a display device of the requester, the search result comprising the record. Other embodiments are described herein.

System and method for using past or external information for future search results

A system and method for integrating business logic into a database system is disclosed herein. One or more processing modules; and one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules and perform: providing a set of distinct records in a database system, the database system comprising a first database cluster H and a second database cluster L, wherein each record of the set of distinct records represents an item for online sale, and wherein the first database cluster H has higher processing capabilities than the second database cluster L; receiving a search request from a requester; comparing the each record of the set of distinct records to a white-list database, wherein one or more records in the white-list database are predicted to be accessed by most requests from other requests within the set of distinct records in the database system; when a first record of the set of distinct records is contained in the white-list database, placing the first record in the first database cluster H; when a second record of the set of distinct records is not contained in the white-list database, calculating a probability that the second record will be accessed and (a) placing the second record of the set of distinct records in the first database cluster H and adding the second record to the white-list database when the second record is likely to be accessed, or (b) placing the second record of the set of distinct records in the second database cluster L when the second record is not likely to be accessed; and presenting a search result containing one or more records of the set of distinct records to the requester. Other embodiments are disclosed herein.

System and method for integrating business logic into a hot/cold prediction

Various embodiments include a system for grouping a set of distinct records in a database system, the database system comprising a first database cluster H and a second database cluster L. In many embodiments, the system can comprise one or more processing modules and one or more non-transitory memory storage modules storing computing instructions configured to run on the one or more processing modules. In some embodiments, the computer instructions can be configured to perform acts of determining, for each distinct record (i) of the set of distinct records, whether the record is a first priority or a second priority; for each distinct record (i) of the set of distinct records which is determined to be the first priority, storing the record in the first database cluster H, wherein the first database cluster H comprises a first computer server; and for each record (i) of the set of distinct records which is determined to be the second priority, storing the record in the second database cluster L. In these embodiments, the second database cluster L can comprise a second computer server different from the first computer server, and the first computer server can have greater processing power than the second computer server. In some embodiments, the first priority can be a higher priority than the second priority. Other embodiments are also disclosed herein.

A system and method for compiling search results is presented herein. A method can include gathering user behavior data, filtering the user behavior data, performing update calculations on the filtered user behavior data, using a decaying function to aggregate the filtered user behavior data; and facilitating the presentation of aggregated user behavior data. The user behavior data can include the user history for a plurality of users at an Internet-accessible site. The site can be an eCommerce site and the user history can include purchase and browsing history that can be used to determine a popularity of items sold by the eCommerce site. Facilitating the presentation can include displaying products in order of popularity. Other embodiments are also disclosed herein.

System for hybrid incremental approach to query processing and method therefor

A system for displaying product information is described herein. The system includes a database and a server including a processor that is programmed to receive search data associated with a product search request received and generate a ranked product list in response to the product search request. The processor selects a first sorting value indicative of a first data category included in the search data, generates a first sorted group of product records as a function of the first sorting value, and stores the first sorted group in the database. The processor selects a second sorting value indicative of a second data category, generates a second sorted group of product records as a function of the first sorted group and the second sorting value, and generates the ranked product list as a function of the second sorted group.

System, method, and non-transitory computer-readable storage media for displaying product information on websites

A system and method for integrating business logic into a database system is disclosed herein. A database system may comprise a first database cluster H and a second database cluster L. After records are separated into the first database cluster H and the second database cluster L, the total number of records in the second database cluster L is compared to a target number of records. If there are too many records in the second database cluster L, a randomly generated number is used to randomly move records from second database cluster L into first database cluster H. In a situation where the first database cluster H is faster and accessed more often, such a method results in more products being placed in the faster database cluster and become more likely to become purchased. Other embodiments are also disclosed herein.

A system and method for searching databases is disclosed herein. A database system may comprise a first database cluster H and a second database cluster L. After receiving a search request from a requester, the search request is executed in the first database cluster H to retrieve a first set of results. If the first set of results is greater than a minimum number of results, then the first set of results are presented to the requester. Otherwise, the second database cluster L is searched to retrieve a second set of results. Then the first set of results and the second set of results are presented to the requester. The first database cluster H is stored on a first database server; and the second database cluster L is stored on a second database server apart from the first database server. Other embodiments are also disclosed herein.

A system and method for adaptive determination of computing resources is presented. Sometimes, computing jobs are to be allocated among a set of computers. Historical information regarding input data size and output data size is retrieved. Capacity information about each machine is retrieved. A weighted function is used to calculate a desired allocation of computing resources based on historical input data size and output data size. Computing resources are allocated based on the calculated allocation. The computing job can be performed on the allocated computing resources. Other embodiments also are disclosed.

System for adaptive determination of computing resources and method therefor

A system and method for using past information for future search results is disclosed. A database system may comprise a first database cluster H and a second database cluster L. Social media mentions are mapped to products in the database system. Thereafter, each record in the database system is tracked to determine if mentions on social media platforms are increasing or decreasing, using a social media popularity trend signal. The social media popularity trend signal is created by comparing social media mentions during one time period to social media mentions in another time period. Other embodiments are also disclosed herein.

Growing community of inventors

Sunnyvale, CA, United States of America

Varun Srivastava