Systems and methods to predict resource availability

1. A method, comprising:

storing, in a database, data for a plurality of retail spaces physically located within one or more building structures and leased or leasable by a plurality of users, wherein the data includes trends for foot traffic within each retail space and sales trends for each retail space;

evaluating, by a computing apparatus, a plurality of pre-defined features, including the trends for foot traffic and the sales trends, for each of the retail spaces based on the data stored in the database to form a dataset of the features for the retail spaces, wherein the dataset is comprised of a training dataset and a test dataset;

applying, by the computing apparatus, a machine learning technique to the training dataset to derive a predictive model that correlates the features for each retail space with a status indicator of whether the retail space was made available during a first predetermined period of time in the past;

applying, by the computing apparatus, the predictive model to the testing dataset to evaluate accuracy of predictions from the predictive model for the testing dataset; and

in response to a determination that the accuracy of predictions from the predictive model for the testing dataset is above a threshold, applying, by the computing apparatus, the predictive model to the dataset to generate predictions of whether the retail spaces of the dataset will be made available during a second predetermined period of time in the future.

2. The method of claim 1 , wherein the retail spaces are in a shopping mall.

3. The method of claim 1 , wherein each retail space's status indicator of whether the retail space was made available during the first predetermined period of time in the past is based on whether a tenant defaulted on leasing the retail.

4. The method of claim 1 , wherein the machine learning technique is a random forest analysis.

5. The method of claim 4 , further comprising:

computing, by the computing apparatus, contributions of the features in the predictive model for making predictions;

ranking, by the computing apparatus, the features based on the contributions of the features; and

providing, by the computing apparatus and based on the ranking, a user interface presenting top contributing features in the predictive model for making predictions.

6. The method of claim 5 , wherein the top contributing features include maximum past due amounts of users leasing the retail spaces.

7. The method of claim 2 , further comprising:

using at least one camera, capturing images within each retail space; and

analyzing the captured images to obtain data pertaining to trends for foot traffic within each retail space.

8. The method of claim 7 , further comprising:

providing a set of robots in the one or more building structures to collect at least a portion of the data stored in the database,

wherein each respective robot in the set of robots includes:

a moving mechanism configured to move the robot on a pathway and within each of the retail spaces that are connected by the pathway;

a controller configured to operate the robot within each of the retail spaces connected by the pathway;

a body having a shape configured to carry items for delivery; and

a camera mounted on the body and controlled by the controller to take images in the area, including within each retail space.

9. The method of claim 8 , wherein the data pertaining to trends for foot traffic data within each of the retail spaces is captured by the robots' cameras.

10. The method of claim 8 , wherein the data pertaining to sales trends for each of the retail spaces is based on items carried by the robots for retailers at the retail space.

11. A non-transitory computer storage media storing instructions configured to instruct a computing apparatus to perform the method, the method comprising:

storing, in a database, data for a plurality of retail spaces physically located within one or more building structures and leased or leasable by a plurality of users, wherein the data includes trends for foot traffic within each retail space and sales trends for each retail space

evaluating, by the computing apparatus, a plurality of pre-defined features, including the trends for foot traffic and the sales trends, for each of the retail spaces based on the data stored in the database to form a dataset of the features for the retail spaces, wherein the dataset is comprised of a training dataset and a test dataset;

applying, by the computing apparatus, a machine learning technique to the training dataset to derive a predictive model that correlates the features for each retail space with a status indicator of whether the retail space was made available during a first predetermined period of time in the past;

applying, by the computing apparatus, the predictive model to the testing dataset to evaluate accuracy of predictions from the predictive model for the testing dataset; and

in response to a determination that the accuracy of predictions from the predictive model for the testing dataset is above a threshold, applying, by the computing apparatus, the predictive model to the dataset to generate predictions of whether the retail spaces of the dataset will be made available during a second predetermined period of time in the future.

12. A computing apparatus, comprising:

at least one microprocessor;

memory storing instructions configured to instruct the at least one microprocessor to at least:

store, in a database, data for a plurality of retail spaces physically located within one or more building structures and leased or leasable by a plurality of users, wherein the data includes trends for foot traffic within each retail space and sales trends for each retail space;

evaluate a plurality of pre-defined features, including the trends for foot traffic and the sales trends, for each of the retail spaces based on the data stored in the database to form a dataset of the features for the retail spaces, wherein the dataset is comprised of a training dataset and a test dataset;

apply a machine learning technique to the training dataset to derive a predictive model that correlates the features for each retail space with a status indicator of whether the retail space was made available during a first predetermined period of time in the past;

apply the predictive model to the testing dataset to evaluate accuracy of predictions from the predictive model for the testing dataset; and

in response to a determination that the accuracy of predictions from the predictive model for the testing dataset is above a threshold, applying, by the computing apparatus, the predictive model to the dataset to generate predictions of whether the retail spaces of the dataset will be made available during a second predetermined period of time in the future.

13. The computing apparatus of claim 12 , wherein the retail spaces are in a shopping mall.

14. The computing apparatus of claim 12 , wherein each retail space's status indicator of whether the retail space was made available during the first predetermined period of time in the past is based on whether a tenant defaulted on leasing the retail.

15. The computing apparatus of claim 12 , wherein the machine learning technique is a random forest analysis.

16. The computing apparatus of claim 15 , wherein the instructions are further configured to instruct the at least one microprocessor to:

compute contributions of the features in the predictive model for making predictions;

rank the features based on the contributions of the features; and

provide, based on the ranking, a user interface presenting top contributing features in the predictive model for making predictions.

17. The computing apparatus of claim 13 , wherein the instructions are further configured to instruct the at least one microprocessor to:

using at least one camera, capture images within each retail space; and

analyze the captured images to obtain data pertaining to trends for foot traffic within each retail space.

18. The computing apparatus of claim 17 , further comprising:

a set of robots in the one or more building structures to collect at least a portion of the data in the database, wherein each respective robot in the set of robots includes:

a moving mechanism configured to move the robot on a pathway and within each of the retail spaces that are connected by the pathway;

a controller configured to operate the robot within each of the retail spaces connected by the pathway;

a body having a shape configured to carry items for delivery; and

a camera mounted on the body and controlled by the controller to take images in the area, including within each retail space.

19. The computing apparatus of claim 18 , wherein the data pertaining to trends for foot traffic data within each of the retail spaces is captured by the robots' cameras.

20. The computing apparatus of claim 18 , wherein the data pertaining to sales trends for each of the retail spaces is based on items carried by the robots for retailers at the retail space.