WO2023146471A2

WO2023146471A2 - Server and method for facilitating marketing for purchasing

Info

Publication number: WO2023146471A2
Application number: PCT/SG2023/050033
Authority: WO
Inventors: Su Hnin Wut YI; Wujuan Lin
Original assignee: Hitachi, Ltd.
Priority date: 2022-01-26
Filing date: 2023-01-18
Publication date: 2023-08-03
Also published as: WO2023146471A3

Abstract

According to various embodiments, there is a server comprising: a processor configured to: extract a geo-location of a mall; refine consumers who visit the mall based on the geo-location of the mall and a location of the consumers; categorise a consumer of the refined consumers into a consumer group of one or more consumer groups; extract online behaviour data relating to purchasing for the consumer group from online behaviour data for each of the one or more consumer groups; extract offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; generate future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group; and generate a marketing strategy for the consumer based on the future behaviour data for the consumer.

Description

SERVER AND METHOD FOR FACILITATING MARKETING FOR PURCHASING

TECHNICAL FIELD

[0001] Various embodiments relate to a server and a method for facilitating marketing for purchasing.

BACKGROUND

[0002] Nowadays, retail industry may be highly competitive. Consumers may shop in a multichannel environment between physical retail stores and digital retail stores. The consumers’ purchasing behaviours may also be changing due to pandemic and other external factors, for example, roadblocks and protests. Attracting and maintaining regular consumers and new consumers may be one of the biggest challenges. The increased use of internet and social media may provide information about the consumers’ purchasing behaviours such as purchasing patterns and reference. Therefore, digital retail stores may provide a targeted marketing promotion prior to sales, since websites may track the consumers’ purchasing behaviours. However, physical retail stores may not be able to provide the targeted marketing promotion, due to lack of consumers’ activities prior to the sales. As a result, the physical retail stores may encounter decreased foot traffic and sales volumes.

[0003] In order to address the problem, information about the consumers’ offline activities relating to purchasing may be used. However, the information about the consumers’ offline activities may not be sufficient to analyse the consumers’ behaviours of purchasing, in order to promote sales volume of retail malls which has been decreased due to low foot traffic data inside the malls and the consumers’ changing behaviours of purchasing pattern from offline to online.

[0004] Therefore, there exists a need to provide an improved solution to integrate the information about the consumers’ offline activities and information about the consumers’ online activities to fully understand the consumers’ behaviours.

SUMMARY [0005] According to various embodiments, there is a server for facilitating marketing for purchasing, the server comprising: a memory for storing instructions; and a processor for executing the stored instructions and configured to: extract a geo-location of a mall; refine consumers who visit the mall based on the geo-location of the mall and a location of the consumers; categorise a consumer of the refined consumers into a consumer group of one or more consumer groups; extract online behaviour data relating to purchasing for the consumer group from online behaviour data for each of the one or more consumer groups; extract offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; generate future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group; and generate a marketing strategy for the consumer based on the future behaviour data for the consumer.

[0006] In some embodiments, the processor is configured to categorise the consumer into the consumer group of the one or more consumer groups based on at least one of a residential area of the consumer and characteristics of the consumer.

[0007] In some embodiments, the processor is configured to: obtain information about online activities relating to purchasing for each of the one or more consumer groups; analyse the information about the online activities relating to purchasing for each of the one or more consumer groups; generate the online behaviour data for each of the one or more consumer groups based on the analysed information about the online activities for each of the one or more consumer groups; obtain information about offline activities relating to purchasing for each of the one or more consumer groups; analyse the information about the offline activities relating to purchasing for each of the one or more consumer groups; and generate the offline behaviour data for each of the one or more consumer groups based on the analysed information about the offline activities for each of the one or more consumer groups.

[0008] In some embodiments, the processor is configured to: correlate the offline behaviour data with the online behaviour data using a spatial-temporal-context parameter; and generate information about a baseline purchasing pattern for the consumer group based on the correlation.

[0009] In some embodiments, the processor is configured to: detect if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for the consumer group; and generate information about a changed purchasing pattern for the consumer group if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern.

[0010] In some embodiments, the processor is configured to generate the marketing strategy for the consumer by combining the information about the baseline purchasing pattern for the consumer group and the information about the changed purchasing pattern for the consumer group.

[0011] In some embodiments, the processor is configured to simulate the marketing strategy before applying the marketing strategy to the consumer.

[0012] In some embodiments, the processor is configured to send a notification relating to the marketing strategy to a device of the consumer.

[0013] In some embodiments, the processor is configured to track the consumer after applying the marketing strategy to the consumer, and use a machine learning algorithm by feeding the marketing strategy and the tracking results as inputs.

[0014] In some embodiments, the processor is configured to match the geo-location of the mall and the location of the consumers, to refine the consumers who visit the mall.

[0015] According to various embodiments, there is a method for facilitating marketing for purchasing, the method comprising: extracting a geo-location of a mall; refining consumers who visit the mall based on the geo-location of the mall and a location of the consumers; categorising a consumer of the refined consumers into a consumer group of one or more consumer groups; extracting online behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; extracting offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; generating future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group; and generating a marketing strategy for the consumer based on the future behaviour data for the consumer.

[0016] In some embodiments, the method further comprises: categorising the consumer into the consumer group of the one or more consumer groups based on at least one of a residential area of the consumer and characteristics of the consumer.

[0017] In some embodiments, the method further comprises: obtaining information about online activities relating to purchasing for each of the one or more consumer groups; analysing the information about the online activities relating to purchasing for each of the one or more consumer groups; generating the online behaviour data for each of the one or more consumer groups based on the analysed information about the online activities for each of the one or more consumer groups; obtaining information about offline activities relating to purchasing for each of the one or more consumer groups; analysing the information about the offline activities relating to purchasing for each of the one or more consumer groups; and generating the offline behaviour data for each of the one or more consumer groups based on the analysed information about the offline activities for each of the one or more consumer groups.

[0018] In some embodiments, the method further comprises: correlating the offline behaviour data with the online behaviour data using a spatial-temporal-context parameter; and generating information about a baseline purchasing pattern for the consumer group based on the correlation.

[0019] In some embodiments, the method further comprises: detecting if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for the consumer group; and generating information about a changed purchasing pattern for the consumer group if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern.

[0020] In some embodiments, the generating the marketing strategy for the consumer comprises generating the marketing strategy for the consumer by combining the information about the baseline purchasing pattern for the consumer group and the information about the changed purchasing pattern for the consumer group.

[0021] In some embodiments, the method further comprises: simulating the marketing strategy before applying the marketing strategy to the consumer.

[0022] In some embodiments, the method further comprises: sending a notification relating to the marketing strategy to a device of the consumer.

[0023] In some embodiments, the method further comprises: tracking the consumer after applying the marketing strategy to the consumer; and using a machine learning algorithm by feeding the marketing strategy and the tracking results as inputs.

[0024] In some embodiments, the refining consumers who visit the mall comprises matching the geo-location of the mall and the location of the consumers, to refine the consumers who visit the mall.

[0025] According to various embodiments, a computer program product, comprising instructions to cause the server of any one of the above embodiments to execute the steps of the method of any one of the above embodiments is provided. [0026] According to various embodiments, a computer-readable medium having stored thereon the above computer program product is provided.

[0027] According to various embodiments, a data processing apparatus configured to perform the method of any one of the above embodiments is provided.

[0028] According to various embodiments, a computer program element comprising program instructions, which, when executed by one or more processors, cause the one or more processors to perform the method of any one of the above embodiments is provided.

[0029] According to various embodiments, a computer-readable medium comprising program instructions, which, when executed by one or more processors, cause the one or more processors to perform the method of any one of the above embodiments is provided. The computer-readable medium may include a non-transitory computer-readable medium.

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In the drawings, like reference characters generally refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the invention. In the following description, various embodiments are described with reference to the following drawings, in which:

[0031] FIG. 1 is a block diagram illustrating a server according to various embodiments.

[0032] FIG. 2 is a block diagram illustrating a processor according to various embodiments.

[0033] FIG. 3 is a conceptual diagram illustrating a system according to various embodiments. [0034] FIG. 4 is a conceptual diagram illustrating a system according to various embodiments. [0035] FIG. 5 is a data flow diagram illustrating a system according to various embodiments. [0036] FIG. 6 is a data flow diagram illustrating a system according to various embodiments. [0037] FIG. 7 is a data flow diagram illustrating a system according to various embodiments.

[0038] FIG. 8 is a data flow diagram illustrating a system according to various embodiments. [0039] FIG. 9 is a data flow diagram illustrating a system according to various embodiments.

[0040] FIG. 10 is a conceptual diagram illustrating a system according to various embodiments.

[0041] FIG. 11 is a conceptual diagram illustrating a system for collecting GPS location of a consumer according to various embodiments.

[0042] FIG. 12 is a data flow diagram illustrating a system according to various embodiments.

[0043] FIG. 13 is a flow diagram illustrating a method according to various embodiments. [0044] FIG. 14 is a flow diagram illustrating a method according to various embodiments.

[0045] FIG. 15 is a flow diagram illustrating a method according to various embodiments.

[0046] FIG. 16 is a flow diagram illustrating a method according to various embodiments.

DESCRIPTION

[0047] Embodiments described below in context of the method are analogously valid for the server, and vice versa. Furthermore, it will be understood that the embodiments described below may be combined, for example, a part of one embodiment may be combined with a part of another embodiment.

[0048] It will be understood that any property described herein for a specific device may also hold for any device described herein. Furthermore, it will be understood that for any device described herein, not necessarily all the components described must be enclosed in the device, but only some (but not all) components may be enclosed.

[0049] It should be understood that the terms “on”, “over”, “top”, “bottom”, “down”, “side”, “back”, “left”, “right”, “front”, “lateral”, “side”, “up”, “down” etc., when used in the following description are used for convenience and to aid understanding of relative positions or directions, and not intended to limit the orientation of any device, structure or any part of any device or structure. In addition, the singular terms “a”, “an”, and “the” include plural references unless context clearly indicates otherwise. Similarly, the word “or” is intended to include “and” unless the context clearly indicates otherwise.

[0050] The term “coupled” (or “connected”) herein may be understood as electrically coupled or as mechanically coupled, for example attached or fixed, or just in contact without any fixation, and it will be understood that both direct coupling or indirect coupling (in other words: coupling without direct contact) may be provided.

[0051] In order that the invention may be readily understood and put into practical effect, various embodiments will now be described by way of examples and not limitations, and with reference to the figures.

[0052] FIG. 1 is a block diagram illustrating a server 1 according to various embodiments.

[0053] In some embodiments, the server 1, for example, implemented by a server computer, may include a communication interface 10, a processor 20, and a memory 30.

[0054] In some embodiments, the memory 30 (also referred to as a “database”) may store input data and/or output data temporarily or permanently. In some embodiments, the memory 30 may store program code which allows the server 1 to perform any one of methods S 100, S200, S300 and S400 (as will be described with reference to FIGS. 14 to 17). In some embodiments, the program code may be embedded in a Software Development Kit (SDK). The memory 30 may include an internal memory of the server 1 and/or an external memory. The external memory may include, but is not limited to, an external storage medium, for example, a memory card, a flash drive, and a web storage.

[0055] In some embodiments, the communication interface 10 may allow one or more external systems to communicate with the processor 20 via a network. In some embodiments, the communication interface 10 may transmit signals to the external systems, and/or receive signals from the external systems via the network.

[0056] In some embodiments, the communication interface 10 may receive information about online activities relating to purchasing for a plurality of consumers from a first external system via the network. In some embodiments, the communication interface 110 may receive information about offline activities relating to purchasing for the plurality of consumers from a second external system via the network. In some embodiments, the first external system and the second external system are different. In some other embodiments, the first external system and the second external system are the same.

[0057] In some embodiments, the online activities relating to purchasing may refer to the consumers’ activities relating to online purchasing/shopping. The information about the online activities relating to purchasing may include, but is not limited to, the consumers’ visit to website, the consumer’s frequencies of pages visited, the consumers’ clicks. In some embodiments, the information about the online activities relating to purchasing may be obtained by a PC and/or a mobile device using advertising ID, IP address, and/or cookies. In some embodiments, the information about the online activities may be collected at a microlevel.

[0058] In some embodiments, the offline activities relating to purchasing may refer to the consumers’ activities relating to offline purchasing/shopping. The information about the offline activities relating to purchasing may include, but is not limited to, consumers’ visit time to malls, consumers’ foot traffic in malls, consumers’ trajectory in the malls, the malls’ geolocation, offline sale data, and/or offline product data. In some embodiments, the information about the offline activities relating to purchasing may be obtained by one or more sensors installed in the malls. For example, the sensor may include a ToF (time-of-flight) sensor. In some embodiments, the information about the offline activities may be collected at a macrolevel.

[0059] In some embodiments, the plurality of consumers may be grouped into one or more consumer groups based on at least one criterion. For example, the plurality of consumers may be grouped into the one or more consumer groups based on at least one of a residential area of the consumers and characteristics of the consumers.

[0060] In some embodiments, the communication interface 10 may receive the information about the online activities relating to purchasing for each of the one or more consumer groups, from the first external system. In some embodiments, the communication interface 10 may receive the information about the offline activities relating to purchasing for each of the one or more consumer groups, from the second external system.

[0061] In some embodiments, the communication interface 10 may allow a plurality of external devices associated with the plurality of consumers, including, but not limited to, a device associated with the a consumer, for example, a device belonging to the consumer, to communicate with the processor 20 via the network. In some embodiments, the communication interface 10 may transmit signals to the external devices, and/or receive signals from the external devices via the network.

[0062] In some embodiments, the communication interface 10 may receive a request for generating a marketing strategy for the consumer. The communication interface 10 may then send the request for generating the marketing strategy to the processor 20.

[0063] In some embodiments, the processor 20 may include, but is not limited to, a microprocessor, an analogue circuit, a digital circuit, a mixed-signal circuit, a logic circuit, an integrated circuit, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), or any combination thereof. Any other kind of implementation of the respective functions, which will be described below in further detail, may also be understood as the processor 20.

[0064] In some embodiments, the processor 20 may be connectable to the communication interface 10. In some embodiments, the processor 20 may be arranged in data or signal communication with the communication interface 10 to receive the request for generating the marketing strategy from the communication interface 10.

[0065] In some embodiments, the processor 20 may extract a geo-location of a mall. For example, the mall may be specified by an owner, a developer, or a manager of the mall. The owner, the developer, or the manager of the mall may initiate a process so that they may first identify the mall and an address of the mall. In some embodiments, the geo-location of the mall may be obtained from the address of the mall.

[0066] In some embodiments, the processor 20 may refine (identify) consumers who visit the mall based on the geo-location of the mall and a location of the consumers. As an example, the consumers who visit the mall may refer to the consumers who usually or regularly visit the mall. In some embodiments, the consumers who usually or regularly visit the mall may be determined based on frequency of the visit to the mall, number of the visit to the mall, etc. In some embodiments, the geo-location of the mall and the location of the consumers may be matched, to refine the consumers as a market segment. In some embodiments, the location of the consumers may include a residential area of the consumers. For example, the location of the consumers may include a GPS location of the consumers. As an example, the GPS location of the consumers may be obtained from a third party advertisement web server (as will be described with reference to FIG. 11). In some embodiments, by refining the consumers, a list of consumers (for example, mobile IDs of the consumers) may be provided. For example, lx people from Simei Road and 2x people from Tampines St 71 may visit a mall A regularly. In this example, the processor 20 may refine the consumers who visit the mall A regularly, and take the lx people and 2x people as target consumers.

[0067] In some embodiments, the processor 20 may receive the information about the online activities relating to purchasing for the plurality of consumers from the communication interface 10. In some embodiments, the processor 20 may receive the information about the offline activities relating to purchasing for the plurality of consumers from the communication interface 10. In some embodiments, the processor 20 may analyse the information about the online activities relating to purchasing for the plurality of consumers, and group the plurality of consumers into the one or more consumer groups, and extract the information about the online activities relating to purchasing for each of the one or more consumer groups. In some embodiments, the processor 20 may analyse the information about the offline activities relating to purchasing for the plurality of consumers, and group the plurality of consumers into the one or more consumer groups, and extract the information about the offline activities relating to purchasing for each of the one or more consumer groups.

[0068] In some other embodiments, the processor 20 may receive the information about the online activities relating to purchasing for each of the one or more consumer groups from the communication interface 10. In some embodiments, the processor 20 may receive the information about the offline activities relating to purchasing for each of the one or more consumer groups from the communication interface 10.

[0069] In some embodiments, the processor 20 may analyse the information about the online activities relating to purchasing for each of the one or more consumer groups. In some embodiments, the processor 20 may generate online behaviour data for each of the one or more consumer groups based on the analysed information about the online activities for each of the one or more consumer groups. In some embodiments, the processor 20 may analyse the information about the offline activities relating to purchasing for each of the one or more consumer groups. In some embodiments, the processor 20 may generate offline behaviour data for each of the one or more consumer groups based on the analysed information about the offline activities for each of the one or more consumer groups.

[0070] In some embodiments, the online behaviour data may refer to data, for example, in the form of a collection of discrete values, that may convey the information about the online activities relating to purchasing. In some embodiments, the offline behaviour data may refer to data, for example, in the form of a collection of discrete values, that may convey the information about the offline activities relating to purchasing.

[0071] In some embodiments, the processor 20 may categorise a consumer of the refined consumers into a consumer group of the one or more consumer groups. In some embodiments, the processor 20 may categorise the consumer into the consumer group of the one or more consumer groups based on at least one criterion. For example, the processor 20 may categorise the consumer into the consumer group of the one or more consumer groups based on at least one of a residential area of the consumer and characteristics of the consumer. For example, the consumer may be tied to the above-mentioned (specified) mall. In some embodiments, the processor 20 may categorise the consumer into one consumer group of the one or more consumer groups. In some other embodiments, the processor 20 may categorise the consumer into two or more consumer groups of a plurality of consumer groups.

[0072] In some embodiments, the processor 20 may extract online behaviour data relating to purchasing for the consumer group that the consumer belongs, from the online behaviour data for each of the one or more consumer groups. In some embodiments, the processor 20 may extract offline behaviour data relating to purchasing for the consumer group, from the offline behaviour data for each of the one or more consumer groups.

[0073] In some embodiments, the processor 20 may predict future behaviour relating to purchasing for the consumer. In some embodiments, the processor 20 may generate future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group that the consumer belongs. In some embodiments, the future behaviour data for the consumer may refer to data, for example, in the form of a collection of discrete values, that may convey information about the consumer’s predicted future activities relating to purchasing.

[0074] In some other embodiments, the processor 20 may correlate the offline behaviour data for the consumer group that the consumer belongs with the online behaviour data for the consumer group, using a spatial-temporal-context parameter. In some embodiments, the processor 20 may generate information about a baseline purchasing pattern for the consumer group based on the correlation. In some other embodiments, the processor 20 may correlate the offline behaviour data for each of the one or more consumer groups with the online behaviour data for each of the one or more consumer groups using a spatial-temporal-context parameter. In some embodiments, the processor 20 may generate information about a baseline purchasing pattern for each of the one or more consumer groups based on the correlation.

[0075] In some embodiments, the baseline purchasing pattern for the consumer group that the consumer belongs may refer to an online/offline purchasing pattern for the consumer group. In some embodiments, the baseline purchasing pattern for each of the one or more consumer groups may refer to an online/offline purchasing pattern for each of the one or more consumer groups. For example, the information about the baseline purchasing pattern is that “a first consumer group (for example, a consumer group of age between 20 and 40 having a baby and living in Tampines area) tends to go to a mall A on a weekend morning to buy a formula”.

[0076] In some embodiments, the processor 20 may detect if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for the consumer group that the consumer belongs. In some embodiments, if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern for the consumer group, the processor 20 may generate information about a changed purchasing pattern for the consumer group. In some other embodiments, the processor 20 may detect if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for each of the one or more consumer groups. In some embodiments, if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern for at least one of the one or more consumer groups, the processor 20 may generate information about a changed purchasing pattern for the at least one of the one or more consumer groups. [0077] In some embodiments, the predetermined key indicator may refer to an indicator which may configure and/or affect the baseline purchasing pattern. For example, the predetermined key indicator may include, but is not limited to, time, day, a product, a mall, and conversion between offline shopping and online shopping. In some embodiments, the information about the changed purchasing pattern may refer to an online/offline purchasing pattern which is changed from the baseline purchasing pattern. For example, the information about the changed purchasing pattern is that “the first consumer group tends to buy the formula from an online shopping mall on a weekday night” or “the first consumer group tends to go to the mall A on a weekend morning to buy a puree”.

[0078] In some embodiments, the processor 20 may generate a marketing strategy for the consumer based on the future behaviour data for the consumer. In some embodiments, the marketing strategy may be to promote sales in the offline shopping, for example, in the mall. For example, the marketing strategy may include, but is not limited to, marketing campaign and targeted sales to the consumer.

[0079] In some embodiments, the processor 20 may generate the marketing strategy for the consumer by combining the information about the baseline purchasing pattern for the consumer group that the consumer belongs and the information about the changed purchasing pattern for the consumer group. The processor 20 may then apply the marking strategy to the consumer. For example, the processor 20 may send a notification relating to the marketing strategy, for example, the marketing campaign or flash sale information, to a device of the consumer.

[0080] In this manner, the server 1 may provide a proactive marketing strategy to the consumers to improve the foot traffic and sales volumes of the malls. This may be achieved by integration of the online behaviour data and the offline behaviour data of the consumers at macro-micro level based on the spatial-temporal-context parameter. The server 1 may detect changes of the foot traffic to the mall and consumers’ behaviours (including purchasing patterns), which may be triggered by external factors such as pandemic, roadblocks, and/or protest. The server 1 may predict a profile of the consumer by categorizing the consumer, and corresponding future behaviour of the consumer based on the consumer group that the consumer belongs. The server 1 may further generate the marketing strategy based on the corresponding future behaviour of the consumer group which is relevant to the predicted profile of the consumer. Therefore, the server 1 may provide efficient and proactive marketing promotions to the right target consumer group at the right time (for example, trigger by changed purchasing pattern) to draw the foot traffic to the mall and generate revenue for the mall. For example, if the first consumer group usually purchased a formula A at the mall A and it is detected that there is increased demands of a product category that the first consumer group usually purchase (for example, the formula) at the online shopping mall and there is decreased demands of the formula at the mall A, the processor 20 may suggest the marketing strategy to recommend a use of vouchers and/or flash sales of similar, related or complemented product (for example, formula B) at the mall A to draw back the foot traffic to the mall A.

[0081] In some embodiments, the processor 20 may simulate the marketing strategy before applying the marketing strategy to the consumer. In some embodiments, the processor 20 may calculate a score for the marketing strategy and apply the marketing strategy to the consumer if the score is equal to or greater than a predetermined value.

[0082] In some embodiments, the processor 20 may track the consumer after applying the marketing strategy to the consumer. In some embodiments, the processor 20 may use a machine learning algorithm by feeding the marketing strategy and the tracking results as inputs. In this manner, the processor 20 may provide an improved marketing strategy to the consumers.

[0083] In some embodiments, the processor 20 may refine the one or more consumer groups based on at least one criterion. In some embodiments, the processor 20 may refine the one or more consumer groups based on at least one of a residential area of the consumers and characteristics of the consumers. In this manner, the consumer groups may be re-grouped considering a changing environment. In some embodiments, data is partitioned by the mall. For example, in a mall A, there may be lx consumers from Simei Road and 2x consumers from Tampines St 71. For the mall A, consumers may be categorized by a residential area of the consumers, and characteristics of the consumers. As another example, in a mall B, there may be lOx consumers from Simei Road and lx consumers from Tampines St 71. For the mall B, consumers may be categorized by a residential area of the consumers, and characteristics of the consumers. In some embodiments, the processor 20 may scale up the above embodiments to a plurality of malls, and re-group the consumer groups based on the plurality of malls (as will be described with reference to FIG. 10).

[0084] FIG. 2 is a block diagram illustrating a processor 20 according to various embodiments. [0085] FIG. 2 illustrates modules of the processor 20 for an integrated retail marketing to determine a correlation of the online behaviour data and the offline behaviour data, to generate the proactive marketing strategy.

[0086] As shown in FIG. 2, the processor 20 may include, but is not limited to, a market segmentation and discovery module 100, an online/offline data extractor 200, a consumers’ online/offline activities integrator 300, a change detection module 350, a marketing strategy generator 400, a scenarios optimisation module 450, and a post-marketing evaluator 500.

[0087] In some embodiments, the market segmentation and discovery module 100 may refine a targeted consumer group (also referred to as a “consumer segment”) based on geo-locations of each mall, consumers’ residence, and other characteristics such as consumers’ visiting frequency to the mall, consumers’ income, and consumers’ lifestyles. For example, the processor 20 may use the market segmentation and discovery module 100 to filter consumers who usually or regularly visit a mall A based on the geo-location of the mall A.

[0088] In some embodiments, the online/offline data extractor 200 may extract the information about the online activities corresponding to the targeted consumer group and the information about the offline activities corresponding to the targeted consumer group. In some embodiments, the online/offline data extractor 200 may include an online data extractor and an offline data extractor, and the online data extractor may extract the information about the online activities corresponding to the targeted consumer group, and the offline data extractor may extract the information about the offline activities corresponding to the targeted consumer group.

[0089] In some embodiments, the consumers’ online/offline activities integrator 300 may correlate the offline behaviour data with the online behaviour data based on macro-level of foot traffic data that visit the mall at the time and micro-level of online consumers’ geo-location data (for example, who visit the mall at the same time), to discover related features between the online behaviour data and the offline behaviour data, in order to generate the baseline purchasing pattern for the consumer group.

[0090] In some embodiments, the change detection module 350 may detect the changed purchasing pattern, when there is deviation of the at least one predetermined key indicator, for example, interests, shopping frequency, and/or spending between online and offline.

[0091] In some embodiments, the marketing strategy generator 400 may generate the marketing strategy for the consumer. In some embodiments, the marketing strategy generator 400 may determine a related consumer profile (also referred to as a “targeted consumer profile”) and products which have a high likelihood of purchases for the targeted consumer profile.

[0092] In some embodiments, the scenarios optimisation module 450 may simulate the marketing strategy by combination of different scenarios for the related consumer profile and distribute messages to consumers having the related consumer profile. [0093] In some embodiments, the post-marketing evaluator 500 may evaluate effects of the marketing strategy based on actual reactions.

[0094] FIG. 3 is a conceptual diagram illustrating a system according to various embodiments. [0095] The system may include a server 1. The server 1 may include the processor 20 including the online/ offline data extractor 200, the consumers’ online/offline activities integrator 300, the change detection module 350, the marketing strategy generator 400, the scenarios optimisation module 450, and the post-marketing evaluator 500.

[0096] In some embodiments, the online/offline data extractor 200 may extract the information about the online activities corresponding to the targeted consumer group. For example, the information about the online activities relating to purchasing may include, but is not limited to, the consumers’ visit to website, the consumer’s frequencies of pages visited, the consumers’ clicks. In some embodiments, the information about the online activities relating to purchasing may be obtained by a PC and/or a mobile device using advertising ID, IP address, and/or cookies. In some embodiments, the information about the online activities may be collected at a micro-level.

[0097] In some embodiments, the online/offline data extractor 200 may extract the information about the offline activities corresponding to the targeted consumer group. For example, the information about the offline activities relating to purchasing may include, but is not limited to, consumers’ visit time to malls, consumers’ foot traffic in malls, consumers’ trajectory in the malls, the malls’ geo-location, offline sale data, and/or offline product data. In some embodiments, the information about the offline activities relating to purchasing may be obtained by one or more sensors installed in the malls. For example, the sensor may include a ToF (time-of-flight) sensor. In some embodiments, the information about the offline activities may be collected at a macro-level.

[0098] In some embodiments, the online behaviour data and the offline behaviour data may be correlated using the spatial-temporal-context parameter. In this manner, macro-scales of the offline foot prints may be associated with micro-scales of the online foot prints. In some embodiments, the correlation may be performed by the consumers’ online/offline activities integrator 300 (also referred to as an “online/offline data integration module”).

[0099] In some embodiments, the change detection module 350 may detect the changed purchasing pattern. For example, the changed purchasing pattern may include, but is not limited to, purchasing cycles, foot traffic, substitutes products, and/or complement products. [00100] In some embodiments, the marketing strategy generator 400 may generate the marketing strategy for the consumer. In some embodiments, the marketing strategy generator 400 may include a complement products discovery module, a related consumer profile discovery module, and the scenario optimisations module 450 (also referred to as a “simulation module”). In some embodiments, the simulation module 450 may simulate the marketing strategy before applying the marketing strategy to the consumer, by combination of different scenarios, for example, tenants, products, similar product price, and/or consumer segments. In some embodiments, the complement products discovery module may identify substitutes and/or complements (accessories, compatible products, etc.) products/product category. In some embodiments, the complement products discovery module may make use of an existing knowledge graph (also referred to as a “products/services category knowledge graph”) which captures the knowledge about products and its related entities in a flexible manner from online and offline. Entities may include objects, things concepts or abstractions that exist (for example, back to school, living room, or minimalist home decoration). It may help to link between products offered in the mall. In some embodiments, the related consumer profile discovery module may identify potential consumer profiles who may have similar attributes (for example, purchasing patterns, purchasing cycles at online and offline).

[00101] In some embodiments, the post-marketing evaluator 500 may include a crossvalidated scoring module and a tracking module. In some embodiments, the Cross-validation scoring module may produce scores for marketing strategy which is generated by an earlier stage. Scoring may be based on benchmarking of strategy efficiency to identify what the best combination of scenarios is at a real-world situation. In some embodiments, the tracking module may generate input data for the cross-validation scoring module by tracking actual effects of the marketing strategy. The tracking module may evaluate the actual effects of the marketing strategy based on actual reactions (for example, by checking whether the consumer visited the mall, whether the consumer purchased the product, whether the consumer purchased or did not purchase at N time frames, whether the consumer attended the event, etc.).

[00102] In some embodiments, the knowledge graph may be used in the online/offline data integration module 300 and the marketing strategy generator 400. In some embodiments, tenant information (for example, merchants’ information in the mall) may be used in the online/offline data integration module 300 and the marketing strategy generator 400. In some embodiments, the knowledge graph may contain information about product offered at online retail sites and offline retail malls. It may be a collection of factual knowledge about products (for example, product types and attribute values). It may particularly provide a source of co- view/complement/substitute products and services categories by querying a specific product type.

[00103] FIG. 4 is a conceptual diagram illustrating a system according to various embodiments.

[00104] The system may include a server 1. The server 1 may include the processor 20 including the market segmentation and discovery module 100, the online/offline data extractor 200, the consumers’ online/offline activities integrator 300, the marketing strategy generator 400, and the post-marketing evaluator 500.

[00105] In some embodiments, the market segmentation and discovery module 100 may refine consumers’ segment based on geo-location of the consumers.

[00106] In some embodiments, the online/offline data extractor 200 may include the online data extractor 210 and the offline data extractor 220. The online data extractor 210 may extract consumers’ online behaviour data via the online activities, for example, consumers’ check-in, cookies, and/or websites. The offline data extractor 220 may extract consumer’s offline behaviour data via offline activities, for example, via various sensors at the mall (or retail stores).

[00107] In some embodiments, the consumers’ online/offline activities integrator 300 may integrate the online behaviour data at micro-level and the offline behaviour data at macro-level. [00108] In some embodiments, the marketing strategy generator 400 may create the marketing strategy, for example, the marketing campaign, an event, and/or a tenant mix scenario (for example, mixing multiple retail stores). In some embodiments, the marketing strategy generator 400 may send relevant notifications to the consumers and/or offers to the consumers to attract to visit the mall (or the retail stores).

[00109] In some embodiments, the post-marketing evaluator 500 may cross-validate the marketing strategy, by tracking effects of the marketing strategy (for example, by checking whether the consumer visited the mall, whether the consumer purchased the product, whether the consumer purchased or did not purchase at N time frames, whether the consumer attended the event, etc.).

[00110] FIG. 5 is a data flow diagram illustrating a system according to various embodiments. FIG. 6 is a data flow diagram illustrating the system according to various embodiments. FIG. 7 is a data flow diagram illustrating the system according to various embodiments. [00111] As shown in FIG. 5, data 105 may be created using a targeted advertising mechanism via an advertising platform, for example, a third party advertisement web server. The advertising platform may have profile data for targeted advertising (also referred to as an “audience”). The audience may be created for each mobile ID and hold information such as gender, age, interests, and geo-location of consumers where they reside, work, and/or visit. The geo-location of visited place may further be linked to geo-location of the malls in order to refine the targeted consumer group who has visited the mall. Data 205 (also referred to as “online data”) may be captured from e-commerce store (also referred to as an “e-store”) or website, using the mobile ID and/or the IP address. The market segmentation and discovery module 100 may further refine targeted consumer segment by geo-location of the mall.

[00112] As shown in FIG. 6, the data 105 may model the consumer characteristics 115, for example, the consumer’s residence location, income, family, timestamp, visiting frequency to the mall, and/or interests at specific time interval. The data 105 may be collected at anonymised macro-level. The data 205 may describe trackable event of the consumer on the internet. The data 205 may include, but is not limited to, shopping frequency at online shopping mall, visited e-commence store, and product category viewed, purchased, searched, reviewed or dropped by the consumer. The online data extractor 210 may produce online consumer behaviour data (or online consumer behaviour model) 215. The data 205 may be collected at anonymised microlevel.

[00113] As shown in FIGS 5 and 7, data 225 (also referred to as “offline data”) may include sensor data 221 which may be extracted from sensors installed inside the mall, and POS transaction data 222 which may be extracted from multiple tenants (retail stores) inside the mall. The offline data extractor 220 may model the information about the offline activities, for example, the timestamp, number of consumers at a point of interest, consumer from-to movement (retail stores to retail stores), dwelling time at visited retail stores, average spending per consumers, frequency per consumers, and product category purchased by consumers. The consumers’ characteristics, for example, demographics, lifestyles, and/or interests, may be mapped with the offline activities at the retail stores. The offline data 225 may further include geo-location of the mall, in order to associate with the online data 110 which may model overall characteristics of the consumer profile.

[00114] FIG. 8 is a data flow diagram illustrating a system according to various embodiments. [00115] FIG. 8 illustrates the data flow diagram of extraction of offline activities for localised consumers through sensors installed inside the mall.

[00116] As shown in FIG. 8, the consumers’ online/offline activities integrator 300 may include a consumers’ behaviour learner module 310, a context factor weighting module 320, and a change detection module 350.

[00117] In some embodiments, the consumers’ behaviour learner module 310 may correlate the online behaviour data and the offline behaviour data using the spatial-temporal-context parameter, wherein who has visited the mall (spatial) at what time (temporal) and which product interested, viewed, and/or purchased along with demographic data (context).

[00118] In some embodiments, the consumers’ behaviour learner module 310 may further learn a context parameter (for example, which product is interested, viewed, and/or purchased at online/offline shopping mall, along with demographic data) by consumer overall characteristics 110, and extract the purchasing patterns. Related, completed, and/or similar product information may be updated to a knowledge graph 330 (also referred to as a “products/services category knowledge graph”) with associated tenant information 340.

[00119] In some embodiments, the context factor weighting module 320 may determine the baseline purchasing pattern derived from the online behaviour data and the offline behaviour data aggregated across different days, time periods, products, and/or brands. The baseline purchasing pattern may reflect prevailing conditions along with necessary calibrations to online and offline data sources.

[00120] In some embodiments, the change detection module 350 may detect the change of the at least one predetermined key indicator, for example, interests, shopping frequency, and/or spending key performance indicators, from the baselines purchasing pattern. The change of the baseline purchasing pattern may then be associated with the consumer profiles as well.

[00121] In some embodiments, output data 370 (also referred to as “consumer profile / interest prediction”) may include the consumer profiles with associated product categories, candidates retail stores, and/or preferred time periods along with changed purchasing pattern to be prioritized.

[00122] As described above and as shown in FIG. 8, product and categories of retail stores may be clustered at the knowledge graph 330. Consumers’ behaviours may be leamt based on the consumer characteristics. When the consumer views and/or purchases items, the consumers’ behaviours may be updated by adding new preferences associated with existing elements along with shopping frequency (online or offline). Consumers’ purchasing patterns may be extracted by learning the consumers’ behaviours. The change may be detected when there is a derivation of preference values and/or shopping frequency at online/offline. The context factor weighting module 320 may incorporate influences of spatial-temporal- contextual factors such as trending products, social closeness, and/or category preference.

[00123] FIG. 9 is a data flow diagram illustrating a system according to various embodiments.

[00124] FIG 9 illustrates the data flow diagram of generating the marketing strategy based on related consumer profiles and products.

[00125] In some embodiments, the marketing strategy generator 400 may include a complement products discovery module 410, a related consumer profile discovery module 415, a different scenarios simulating module 420, and the scenarios optimisation module 450.

[00126] In some embodiments, the complement products discovery module 410 may search candidate products of consumer profiles data 370, and discover new complement and/or similar products from the knowledge graph 330. For example, a consumer A may buy skin care products together with hair care products. In this example, the complement products discovery module 410 may identify the skin care products and the hair care products.

[00127] In some embodiments, in the related consumer profile discovery module 415, the consumer profiles data 370 may be used to discover similar consumer profile in the online/offline shopping mall. For example, consumer profile A always buys the skin care products and the hair care products at the online shopping mall, but the consumer A buys shoes at the offline shopping mall. Another consumer profile B has similar characteristics with the consumer profile A such as age, gender, lifestyles, incomes, and preference. In this example, the related consumer profile discovery module 415 may identify that the consumer profile A and the consumer profile B are related.

[00128] In some embodiments, the different scenarios simulating module 420 may run combinations of multiple “what-if’ scenarios across discovered products, brands and/or discovered consumer profiles to see the simulated effects of such combinations. Each simulation exercise may create a list of recommendation strategies 435, and propose timings, complement and/or similar products, and/or brands for each consumer profile based on scores from cross validation 505.

[00129] In some embodiments, as shown in the box 490, the marketing strategy may be realised. The marketing strategy may include event campaigns or distribution of messages to consumers to attract to the offline retail mall. [00130] In some embodiments, the post-marketing evaluator 500 may feedback postmarketing campaign evaluation to the simulation model in order to improve accuracy of recommendations and to create effective marketing strategy.

[00131] In some embodiments, inputting the consumer profiles data 370 may include a group of consumer segments who has less frequency to visit the mall and more frequency to visit the online shopping mall. The inputting the consumer profiles data 370 may also include a group of consumers who has similar interests and/or purchasing patterns with a previous group but has never visited the mall.

[00132] FIG. 10 is a conceptual diagram illustrating a system according to various embodiments.

[00133] FIG. 10 is the conceptual diagram illustrating a view of multiple retail malls which is distributed across multiple locations.

[00134] For example, multiple geo-location of the malls may be able to scale the system for multiple sets of localised consumers’ online/offline purchasing behaviours analysis. The system may scale the solution according to various embodiments, by setting geo-location for each mall under the same mall operator.

[00135] FIG. 11 is a conceptual diagram illustrating a system for collecting GPS location of a consumer according to various embodiments.

[00136] As shown in FIG. 11, data may be created using a targeted advertising mechanism via an advertising platform, for example, a third party advertisement web server. The advertising platform may have profile data for targeted advertising (also referred to as an “audience”). The audience may be created for each mobile ID and hold information such as gender, age, interests, and geo-location (for example, GPS) of consumers where they reside, work, and/or visit. The geo-location of visited place may further be linked to geo-location of the malls in order to refine the targeted consumer group who has visited the mall.

[00137] As shown in FIG. 11, in some embodiments, to collect the data, a trade area may be created. Thereafter, time period for the collection of the data may be specified. Thereafter, the data may be automatically collected within the trade area.

[00138] FIG. 12 is a data flow diagram illustrating a system according to various embodiments.

[00139] FIG. 12 is the data flow diagram illustrating the system showing high level functional steps. [00140] As shown in FIG. 12, in some embodiments, offline data may include sensor data which may be extracted from sensors installed inside the mall, and POS transaction data which may be extracted from multiple tenants (retail stores) inside the mall. In some embodiments, omnichannel transaction data may be collected using the POS transaction data and online transaction data. In some embodiments, location data of the retail stores and the malls may be collected.

[00141] In some embodiments, social trends data may be collected.

[00142] In some embodiments, online data may be captured from the e-commerce store (also referred to as an “e-store”) or website, using the mobile ID and/or the IP address.

[00143] In some embodiments, consumers’ lifestyle data, for example, the consumer’s residence location, income, family, timestamp, visiting frequency to the mall, and/or interests at specific time interval, may be collected.

[00144] In some embodiments, the offline data extractor 220 may model the information about the offline activities, for example, the timestamp, number of consumers at a point of interest, consumer from-to movement (retail stores to retail stores), dwelling time at visited retail stores, average spending per consumers, frequency per consumers, and product category purchased by consumers.

[00145] In some embodiments, the online data extractor 210 may model the information about the online activities, for example, timestamp, shopping frequency at online shopping mall, visited shopping category/merchants, and similar and/or related product category viewed, purchased, searched, reviewed or dropped by the consumer.

[00146] In some embodiments, a deep learning model may be used to discover related features from the online behaviour data and the offline behaviour data.

[00147] In some embodiments, the consumer profile/interest prediction model 370 may include the consumer profiles with associated product categories, candidates retail stores, and/or preferred time periods (date/time/week) along with changed purchasing pattern to be prioritized. In some embodiments, consumer characteristics and preference may be predicted based on the spatial-temporal-context parameter, and similar and/or related product relationship.

[00148] In some embodiments, the marketing strategy may be created. For example, a recommendation of the marketing strategy may be generated. The recommended marketing strategy may be scored from cross-validation. [00149] In some embodiments, the marketing strategy may be applied, for example, by sending the notification to the device of the consumer. In some embodiments, after sending the notification, the consumer may be tracked. Based on the tracking, consumer’s purchase data may be collected and conversion rate (for example, successful rate of the marketing strategy) may be calculated. In some embodiments, cross-validation may be performed for reinforcement learning. The score may be updated based on whether the consumer purchased the product after sending the notification.

[00150] FIG. 13 is a flow diagram illustrating a method S100 according to various embodiments.

[00151] In some embodiments, the method S100 may include a step S101 of extracting a geo-location of a mall. For example, the mall may be specified by an owner, a developer, or a manager of the mall. In some embodiments, the geo-location of the mall may be obtained from an address of the mall.

[00152] In some embodiments, the method S100 may include a step S102 of refining consumers who visit the mall based on the geo-location of the mall and a location of the consumers. As an example, the consumers who visit the mall may refer to the consumers who usually or regularly visit the mall. In some embodiments, the geo-location of the mall and the location of the consumers may be matched, to refine the consumers as a consumers’ segment (also referred to as a “market segment”). As an example, the location of the consumers, for example, a GPS location of the consumers, may be obtained from a third party advertisement web server (as described with reference to FIG. 11).

[00153] In some embodiments, the method S100 may include a step S103 of categorising a consumer of the refined consumers into a consumer group of one or more consumer groups. In some embodiments, each consumer of the refined consumers may be categorised into a respective consumer group of the one or more consumer groups.

[00154] In some embodiments, the method S100 may include a step S104 of extracting online behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups. For example, the online behaviour data may be obtained from website.

[00155] In some embodiments, the method S100 may include a step S105 of extracting offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups. For example, the offline behaviour data may be obtained from the mall. [00156] In some embodiments, the method S100 may include a step S106 of generating future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group.

[00157] In some embodiments, the method S100 may include a step S107 of generating a marketing strategy for the consumer based on the future behaviour data for the consumer.

[00158] FIG. 14 is a flow diagram illustrating a method S200 according to various embodiments.

[00159] In some embodiments, the method S200 may include a step S201 of refining targeted consumers based on geographic segments (for example, residence locations and/or mall locations).

[00160] In some embodiments, the method S200 may include a step S202 of extracting online behaviour data and offline behaviour data for the targeted groups.

[00161] In some embodiments, the method S200 may include a step S203 of determining a spatial-temporal-context parameter to integrate the online behaviour data and the offline behaviour data.

[00162] In some embodiments, the method S200 may include a step S204 of determining a baseline purchasing pattern.

[00163] In some embodiments, the method S200 may include a step S205 of detecting a changing purchasing pattern.

[00164] In some embodiments, the method S200 may include a step S206 of generating a proactive marketing strategy based on related consumer characteristics and similar and/or complemented product relationships.

[00165] In some embodiments, the method S200 may include a step S207 of feedback by post-marketing campaign evaluation.

[00166] FIG. 15 is a flow diagram illustrating a method S300 performed by the system according to various embodiments.

[00167] In some embodiments, the method S300 may include a step S301 of refining targeted consumers segments based on geo-locations of each mall, consumer’s residence, and other characteristics such as visiting frequency to the mall, income, and/or lifestyles. In the step S301, the system may extract online activities for the targeted consumers followed by refining the targeted consumers based on geographic segments. The system may define where the consumers reside, where the consumers work, and/or where the consumers visit along with demographic data such as incomes, lifestyles, age, and/or gender. [00168] In some embodiments, the method S300 may include a step S302 of extracting the online behaviour data relating to the online activities which may include any measurable or trackable event of a user (a consumer) on the internet for each targeted consumer segment. The online activities may include, but is not limited to, visits to websites, and/or frequency of pages visited along with tracking what product is viewed and/or purchased.

[00169] In some embodiments, the method S300 may include a step S3O3 of extracting the offline behaviour data relating to the offline activities collected from sensors which are installed inside the mall. The offline data may include, but is not limited to, total foot traffic of the mall, and/or movement information of consumers inside the mall over the same period at specific interval. In the step S3O3, the offline activities of the consumers at the retail malls may be captured. The offline activities may include, but is not limited to, consumers’ movement inside the mall, total foot traffic, visited shops, and/or dwelling time at anonymized level.

[00170] In some embodiments, the method S300 may include a step S304 of correlating the online behaviour data and the offline behaviour data using the spatial-temporal-context parameter, wherein who has visited the mall (spatial) at what time (temporal) and which product interested, viewed, and/or purchased along with demographic data (context). For example, there are total 100 online consumers from residence X who visited the mall A at 10 am, another 20 online consumers from residence Y who visited the mall A at the same time which is captured from online data through users’ (consumers’) mobile GPS location. On the other hand, there are total 120-foot traffic at mall A at 10 am which is captured from offline data via sensors installed inside the mall. These 100 online consumers are associated with 100 out of 120 offline data inside the mall and another 20 online consumers are also associated with 20 out of 120 offline data. Therefore, the system may associate the offline data (for example, movement of consumers, dwelling time, and/or visited shops) with the online data (for example, visits to websites, frequency of pages, product views, product purchases, and/or purchasing cycle) .

[00171] In some embodiments, the method S300 may include a step S305 of determining baseline purchasing pattern derived from the online behaviour data and the offline behaviour data aggregated across different days, time periods, products, and/or brands. The baseline purchasing pattern may reflect prevailing conditions along with necessary calibrations to data sources. The change may be detected when there is deviation of interests, shopping frequency, and/or spending key performance indicators from the baseline purchasing pattern. [00172] In some embodiments, the method S300 may include a step S306 of generating a marking strategy for the mall by combining the baseline purchasing pattern and the changed purchasing pattern which has been determined in earlier stage. The marketing strategy may include discovery of related or complemented products or brands for similar consumer profiles. [00173] In some embodiments, the method S300 may include a step S307 of determining optimal configuration combinations based on newly discovered related or complemented products or brands for the similar consumer profiles. In the step S307, the system may simulate the marketing strategy, by combination of different scenarios for related consumer profiles and distribute messages to consumers. The system may determine the optimal combinations of products, brands, and/or consumer profiles along with scores from evaluations by the feedback from post-marketing campaigns 500.

[00174] In some embodiments, the method S300 may include a step S3O8 of feedback by the post-marketing campaign evaluation to the system using data acquired from actual tracking of consumer profiles reactions by comparing key performance indicators such as increase in foot traffic and increase in sales volumes.

[00175] FIG. 16 is a flow diagram illustrating a method S400 according to various embodiments.

[00176] FIG. 16 illustrates the method S400 of determining changing behaviours of consumers from correlation of the online behaviour data and the offline behaviour data in order to generate consumer profile with associated product categories which will highly have a likelihood of purchasing at the retail stores. The profile may further indicate a candidate list of products, brands, and/or preference to be targeted by the marketing strategy generator 400.

[00177] In some embodiments, the method S400 may include a step S401 of clustering product and/or categories of retail stores to form product, services, and/or categories knowledge graph, and acquiring retail stores’ information with products, services, and/or categories which is offered in the mall to form tenant information.

[00178] In some embodiments, the method S400 may include a step S402 of correlating the online behaviour data and the offline behaviour data using the spatial-temporal-context parameter, wherein who has visited the mall (spatial) at what time (temporal) and which product interested, viewed, purchased along with demographic data (context).

[00179] In some embodiments, the method S400 may include a step S403 of learning the context parameter (for example, which product is interested, viewed, and/or purchased at online/offline shopping mall along with demographic data) by consumers’ overall characteristics and extracting the purchasing pattern. Related, completed, and/or similar product information may be updated to the knowledge graph with associated tenant information.

[00180] In some embodiments, the method S400 may include a step S404 of determining the baseline purchasing pattern derived from the online behaviour data and the offline behaviour data aggregated across different days, time periods, products, and/or brands. The baseline purchasing pattern may reflect the prevailing conditions along with necessary calibrations to online and offline data sources.

[00181] In some embodiments, the method S400 may include a step S405 of detecting the change of the purchasing pattern, when there is deviation of interests, shopping frequency, and/or spending key performance indicators from the baseline purchasing pattern. The change of the purchasing pattern may then be associated with consumer profiles as well.

[00182] In some embodiments, the method S400 may include a step S406 of generating the consumer profile with associated product categories, candidates retail stores, and/or preferred time periods along with changing purchasing pattern to be prioritized.

[00183] As described above, according to various embodiments, the system may eliminate the need for individual targeting and related privacy concern for shoppers, yet it detects macro phenomena of changing behaviours from micro-level. The system may provide the proactive marketing strategy by simulating different scenarios in order to attract new shoppers as well as regular shoppers.

[00184] While embodiments of the invention have been particularly shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. The scope of the invention is thus indicated by the appended claims and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced. It will be appreciated that common numerals, used in the relevant drawings, refer to components that serve a similar or the same purpose. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. Unless specifically stated otherwise, the term “some” refers to one or more. Combinations such as “at least one of A, B, or C,” “one or more of A, B, or C,” “at least one of A, B, and C,” “one or more of A, B, and C,” and “A, B, C, or any combination thereof’ include any combination of A, B, and/or C, and may include multiples of A, multiples of B, or multiples of C. Specifically, combinations such as “at least one of A, B, or C,” “one or more of A, B, or C,” “at least one of A, B, and C,” “one or more of A, B, and C,” and “A, B, C, or any combination thereof’ may be A only, B only, C only, A and B, A and C, B and C, or A and B and C, where any such combinations may contain one or more member or members of A, B, or C. All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. The words “module,” “mechanism,” “element,” “device,” and the like may not be a substitute for the word “means.” As such, no claim element is to be construed as a means plus function unless the element is expressly recited using the phrase “means for.”

Claims

1. A server for facilitating marketing for purchasing, the server comprising: a memory for storing instructions; and a processor for executing the stored instructions and configured to: extract a geo-location of a mall; refine consumers who visit the mall based on the geo-location of the mall and a location of the consumers; categorise a consumer of the refined consumers into a consumer group of one or more consumer groups; extract online behaviour data relating to purchasing for the consumer group from online behaviour data for each of the one or more consumer groups; extract offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; generate future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group; and generate a marketing strategy for the consumer based on the future behaviour data for the consumer.

2. The server according to claim 1, wherein the processor is configured to categorise the consumer into the consumer group of the one or more consumer groups based on at least one of a residential area of the consumer and characteristics of the consumer.

3. The server according to claim 1 or claim 2, wherein the processor is configured to: obtain information about online activities relating to purchasing for each of the one or more consumer groups; analyse the information about the online activities relating to purchasing for each of the one or more consumer groups; generate the online behaviour data for each of the one or more consumer groups based on the analysed information about the online activities for each of the one or more consumer groups; obtain information about offline activities relating to purchasing for each of the one or more consumer groups; analyse the information about the offline activities relating to purchasing for each of the one or more consumer groups; and generate the offline behaviour data for each of the one or more consumer groups based on the analysed information about the offline activities for each of the one or more consumer groups.

4. The server according to claim 3, wherein the processor is configured to: correlate the offline behaviour data with the online behaviour data using a spatial- temporal-context parameter; and generate information about a baseline purchasing pattern for the consumer group based on the correlation.

5. The server according to claim 4, wherein the processor is configured to: detect if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for the consumer group; and generate information about a changed purchasing pattern for the consumer group if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern.

6. The server according to claim 5, wherein the processor is configured to generate the marketing strategy for the consumer by combining the information about the baseline purchasing pattern for the consumer group and the information about the changed purchasing pattern for the consumer group.

7. The server according to any one of claims 1 to 6, wherein the processor is configured to simulate the marketing strategy before applying the marketing strategy to the consumer.

8. The server according to any one of claims 1 to 7, wherein the processor is configured to send a notification relating to the marketing strategy to a device of the consumer.

9. The server according to any one of claims 1 to 8, wherein the processor is configured to track the consumer after applying the marketing strategy to the consumer, and use a machine learning algorithm by feeding the marketing strategy and the tracking results as inputs.

10. The server according to claim 1, wherein the processor is configured to match the geolocation of the mall and the location of the consumers, to refine the consumers who visit the mall.

11. A method for facilitating marketing for purchasing, the method comprising: extracting a geo-location of a mall; refining consumers who visit the mall based on the geo -location of the mall and a location of the consumers; categorising a consumer of the refined consumers into a consumer group of one or more consumer groups; extracting online behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; extracting offline behaviour data relating to purchasing for the consumer group from offline behaviour data for each of the one or more consumer groups; generating future behaviour data relating to purchasing for the consumer based on the online behaviour data and the offline behaviour data for the consumer group; and generating a marketing strategy for the consumer based on the future behaviour data for the consumer.

12. The method according to claim 11 further comprising categorising the consumer into the consumer group of the one or more consumer groups based on at least one of a residential area of the consumer and characteristics of the consumer.

13. The method according to claim 11 or claim 12 further comprising: obtaining information about online activities relating to purchasing for each of the one or more consumer groups; analysing the information about the online activities relating to purchasing for each of the one or more consumer groups; generating the online behaviour data for each of the one or more consumer groups based on the analysed information about the online activities for each of the one or more consumer groups; obtaining information about offline activities relating to purchasing for each of the one or more consumer groups; analysing the information about the offline activities relating to purchasing for each of the one or more consumer groups; and generating the offline behaviour data for each of the one or more consumer groups based on the analysed information about the offline activities for each of the one or more consumer groups.

14. The method according to claim 13 further comprising: correlating the offline behaviour data with the online behaviour data using a spatial- temporal-context parameter; and generating information about a baseline purchasing pattern for the consumer group based on the correlation.

15. The method according to claim 14 further comprising: detecting if there is a deviation of at least one predetermined key indicator from the baseline purchasing pattern for the consumer group; and generating information about a changed purchasing pattern for the consumer group if is it detected that there is the deviation of the at least one predetermined key indicator from the baseline purchasing pattern.

16. The method according to claim 15, wherein the generating the marketing strategy for the consumer comprises generating the marketing strategy for the consumer by combining the information about the baseline purchasing pattern for the consumer group and the information about the changed purchasing pattern for the consumer group.

17. The method according to any one of claims 11 to 16 further comprising simulating the marketing strategy before applying the marketing strategy to the consumer.

18. The method according to any one of claims 11 to 17 further comprising sending a notification relating to the marketing strategy to a device of the consumer.

19. The method according to any one of claims 11 to 18 further comprising: tracking the consumer after applying the marketing strategy to the consumer; and using a machine learning algorithm by feeding the marketing strategy and the tracking results as inputs.

20. The method according to claim 11, wherein the refining consumers who visit the mall comprises matching the geo-location of the mall and the location of the consumers, to refine the consumers who visit the mall.