Estimating footfall for a location

A system for estimating footfall for a location of one or more users, by a processor and memory circuitry (PMC), the processor being configured to:

• a) obtain visitation data including a plurality of visit data items, wherein each visit data item includes at least data indicative of visit time and visit coordinates associated with a user of the one or more users;
• b) group the obtained visitation data into one or more groups of visit data items, wherein each group is associated with a respective user;
• c) for each group, assign the data items in the group to one or more clusters of visits, including, for each visit data item:
• assign the data item to a given cluster of the clusters if at least the visit time data and the visit coordinates data of the data item comply with a corresponding visit time and visit coordinates conditions that are associated with the given cluster, and
• d) determine, based on the one or more clusters of visits data, a footfall estimation for a location, by:
• selecting at least one machine learning (ML) estimation model of the plurality of ML estimation models, from a plurality of ML estimation models, wherein the selected ML estimation model is trained to estimate footfall of a location while operating on a set of features, wherein at least one feature of the set of features is a feature relating to a cluster of visits;
• wherein the selection is according to at least one feature of the operated set of features;
• and
• using the selected ML estimation model for determining the footfall estimation for one or more locations;
• wherein the footfall estimation for the location is indicative of a number of visits of the one or more users in the location within one or more time intervals.

A non-transitory computer readable storage medium tangibly embodying a program of instructions that, when executed by a computer, cause the computer to perform a method for estimating a footfall for a location of one or more users, by a processor and memory circuitry (PMC), the method comprising:

• a) obtaining visitation data including a plurality of visit data items, wherein each visit data item includes at least data indicative of visit time and visit coordinates associated with a user of the one or more users;
• b) grouping the obtained visitation data into one or more groups of visit data items, wherein each group is associated with a respective user;
• c) for each group, assigning the data items in the group to one or more clusters of visits, including, for each visit data item:
• assigning the data item to a given cluster of the clusters, if at least the visit time data and the visit coordinates data of the data item comply with corresponding visit time and visit coordinates conditions that are associated with the given cluster, and
• d) determining, based on one or more clusters of visits data, a footfall estimation of a location, by:
• selecting at least one machine learning (ML) estimation model of the plurality of ML estimation models, from a plurality of ML estimation models, wherein the selected ML estimation model is trained to estimate footfall of a location while operating on a set of features, wherein at least one feature of the set of features is a feature relating to a cluster of visits;
• wherein the selection is according to at least one feature of the operated set of features;
• and
• using the selected ML estimation model for determining the footfall estimation for one or more locations;
• wherein the footfall estimation for the location is indicative of a number of visits of the one or more users in the one or more locations in one or more time intervals.

A computerized method for training a machine learning (ML) estimation model to estimate footfall for a location, by a processor and memory circuitry (PMC), the computerized method comprising:

• a) obtaining training visitation data including a plurality of visit data items, wherein each visit data item includes at least data indicative of visit time and visit coordinates associated with a user;
• b) grouping the obtained training visitation data into one or more groups of visit data items, wherein each group is associated with a respective user;
• c) for each group, assigning the data items in the group to one or more clusters of visits data, including, for each data item:
• assigning the data item to a given cluster of the clusters if at least the visit time data and the visit coordinates data of the data item comply with corresponding visit time and visit coordinates conditions that are associated with the given cluster,
• d) training a machine learning (ML) estimation model having a specified set of features, based on one or more clusters of visits data, to estimate footfall for a given location, wherein the footfall estimation is indicative of a number of visits of one or more users in the location within one or more time intervals;
• wherein the trained ML estimation models are used for estimating footfall for a location of one or more users, after obtaining visitation data, grouping the obtained visitation data into one or more groups of visit data items, for each group, assigning the data items in the group to one or more clusters of visits, and determining, based on the one or more clusters of visits data, a footfall estimation for a location.

A system for estimating footfall for training a machine learning (ML) estimation model to estimate footfall for a location, by a processor and memory circuitry (PMC), the processor configured to execute a method comprising:

• a) obtaining training visitation data including a plurality of visit data items, wherein each visit data item includes at least data indicative of visit time and visit coordinates associated with a user;
• b) grouping the obtained training visitation data into one or more groups of visit data items, wherein each group is associated with a respective user;
• c) for each group, assigning the data items in the group to one or more clusters of visits data, including, for each data item:
• assigning the data item to a given cluster of the clusters if at least the visit time data and the visit coordinates data of the data item comply with corresponding visit time and visit coordinates conditions that are associated with the given cluster,
• d) training a machine learning (ML) estimation model having a specified set of features, based on one or more clusters of visits data, to estimate footfall for a given location, wherein the footfall estimation is indicative of a number of visits of one or more users in the location within one or more time intervals;
• wherein the trained ML estimation models are used for estimating footfall for a location of one or more users, after obtaining visitation data, grouping the obtained visitation data into one or more groups of visit data items, for each group, assigning the data items in the group to one or more clusters of visits, and determining, based on the one or more clusters of visits data, a footfall estimation for a location.