WO2011026121A2

WO2011026121A2 - Systems, methods, and computer program products for user identification in communication networks

Info

Publication number: WO2011026121A2
Application number: PCT/US2010/047362
Authority: WO
Inventors: Ariel Fligler
Original assignee: Connectiva Systems
Priority date: 2009-08-31
Filing date: 2010-08-31
Publication date: 2011-03-03
Also published as: WO2011026121A3

Abstract

A method and system for determining an identity of a user over a network that may not otherwise provide for user identification as such, e.g., a pre-paid mobile telephone user. According to an embodiment of the invention, a plurality of user profiles may be compiled based on network communication usage, and for new users, a usage profile may be compiled and compared to previously complied user profiles. Based on a degree of similarity, a determination may be made that the user corresponds to a user that previously used the system. Various usage profiles may be used, including, e.g., social network, geographic usage, account activity, etc.

Description

SYSTEMS, METHODS, AND COMPUTER PROGRAM PRODUCTS FOR USER IDENTIFICATION IN COMMUNICATION NETWORKS

BACKGROUND OF THE INVENTION

Customers using mobile devices and services, e.g., cellular phones, can typically consume communication services based on two main modes: post-paid and pre-paid. In post-paid service, the consumer is billed by the end of the month after it has used the services. Such billing scheme is possible with people with credit lines, with which the operator can build long lasting relationships. In pre-paid schemes, the user buys a service usage allowance amount in advance, and each usage is debited from this credit.

Pre-paid is popular with people with no credit lines such as teens, tourists, or foreign citizens. Pre-paid schemes are also popular due to some advantages, including: (a) in many countries, pre-paid phone usage may be had without disclosing personal details, thereby enabling anonymity of the user; and (b) as the customer is not bounded by long term usage agreements, he has the freedom to move between operators, aiming to get the most lucrative plan available at any time.

In some markets, for example, Italy, and certain developing countries, as many as 80-90% of mobile customers use pre-paid schemes. In addition, several trends further strengthen the popularity of the pre-paid billing usage: (a) penetration strategies of mobile virtual network operators (MVNOs) are increasingly based on pre-paid users; (b) economic recessions may provide incentives for keeping mobile service, but containing recurring costs; and (c) handset evolution is happening at a faster pace, and post-paid plans with two-year contracts lock customers in to specific devices; however, pre-paid plans allow faster handset replacement cycles.

In general, pre-paid billing schemes present several challenges to operators. First, as customers are anonymous, operators have a problem developing and managing relationship with them. Further, with no address, anonymous customers are also harder to reach. It is also harder transforming those customers into post-paid users.

Second, prepaid users have low commitment to the operator and the result is high churn rates. Many times, price wars between operators hurt the profit margins.

Lastly, mobile service providers measure their performance in part by subscriber counts. With prepaid service, reporting is more complicated, affecting public company share values and increasing uncertainty, unless these pre-paid users can be accurately measured.

SUMMARY OF EMBODIMENTS OF THE INVENTION

According to various embodiments of the invention, the herein proposed systems and method may enable the user to identify customers along their appearance in the network (even with different identities). This enables the operator to develop marketing relationships with pre-paid customers, further an operator to (a) offer promotions and incentives, (b) understand customer usage patterns (since now the user's different identities are aggregated into a single identity along time, his usage patterns are more accurate), (c) provide better personalized experience, even when the customer resurfaces in the network with new identity, the operator can fetch his profile and provide him with a customized experience (as consumption patterns are known). It will be recognized that the present invention may be used to increase loyalty and higher ARPU of the pre-paid user. This translates later to higher probability that the high ARPU prepaid customer will transform to a post-paid one.

Next, the present invention may be used to accurately identify the high ARPU customers. Economical models have shown that it is a better strategy to keep the high ARPU customers, rather than invest in acquiring new customers.

One of the most problematic areas of marketing in the pre-paid domain is the one of price wars. With low margins, operators have little flexibility of making price wars though pre-paid customers are mainly sensitive to the price. Having the ability to select a fine tuned group of customers and approach them directly and personally, with selected promotion, allows the operator to make personalized price wars (namely offer a promotion without the knowledge of the competing operator as the promotion is below the line marketing event) to keep his cherished customers in house.

The present invention may be used to allow identification of post-paid users when they are using pre-paid accounts. For example, people may buy a pre-paid card when they want to have a specific anonymous usage. Providing benefits in this channel as well, helps strengthening the customer's loyalty to the brand.

Since customers' usage of the network is accurate, the operator has accurate churn and active users statistics; some customers change identity rather than churn while remerging using new identities. Since active users' information is influencing the valuation of mobile operators and their share's prices, accurate churn information is of uttermost importance.

The present invention may further assist in detection and resolution of identify theft. Although fraud is not the main purpose of the present system, if an identified number resurfaces with a significantly different usage pattern, the system can point that fact to the operator, to mitigate a risk of fraud.

According to some embodiments of the invention, the herein disclosed systems and methods may enable network operators (e.g., a mobile operator) or associated entities (e.g., marketers) to develop marketing relationships with their anonymous pre-paid users, by identifying the customers along their usage of the network even if they are appearing with different identities (MSISDN numbers). This may be done by capturing the customer's unique social, usage habits and/or geographic fingerprints. By this identification, operators can provide customized promotions and personalized services, increasing ARPU and loyalty in the competitive pre-paid market.

BRIEF DESCRIPTION OF THE DRAWINGS

The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings in which:

Fig. 1 is a schematic flow diagram showing a method according to an embodiment of the present invention;

Fig. 2 is a block diagram of user identification as may be implemented in systems and methods according to embodiments of the invention;

Fig. 3 is a schematic diagram of a state machine that may be implemented as a system or method for identifying users of a network according to embodiments of the present invention;

Fig. 4 is a schematic illustration of a system for user identification according to embodiments of the invention;

Fig. 5 is a schematic diagram of a batch use case according to an embodiment of the invention; and

Fig. 6 is a schematic diagram of an online pull use case according to an embodiment of the invention.

It will be appreciated that for simplicity and clarity of illustration, elements shown in the figures have not necessarily been drawn to scale. For example, the dimensions of some of the elements may be exaggerated relative to other elements for clarity. Further, where considered appropriate, reference numerals may be repeated among the figures to indicate corresponding or analogous elements.

DESCRIPTION OF THE PRESENT INVENTION

The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings.

In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, and components have not been described in detail so as not to obscure the present invention.

The herein disclosed systems and methods may address and handle situations in which a prepaid user may appear and reappear in the network using different identities (e.g., different MSISDN numbers and/or different mobile devices). When a pre-paid customer buys a new prepaid phone, he may be assigned a different number (i.e., a number with which the user was not previously associated in the system), and thus, without more information, the operator cannot detect that the user is the same customer as had previously used a different number. The proposed systems and methods enable operators of communication networks (e.g., cellular telephony, WiFi, WiMax, etc.) to identify a pre-paid customer even despite having changed an identifying number, e.g., MSISDN. A service may be provided to the user based on this identification.

The herein disclosed systems and methods implement identification of one or more identifying patterns of a user (which is usually pre-paid user, but it is clear that the invention can be extended to other types of network users as well - e.g. non-paying users, and even rough unauthorized users), also known as fingerprints. The fingerprints of each user are substantially unique to that user, and may be valid regardless of the identity being used by the user for communicating over the network.

Some such identification patterns (fingerprints) which may be implemented in various embodiments of the invention are: (a) the people/entities with which the user is in relationship (which may be other users of the network, but also external to the network - e.g. land telephony user, internet users, and so forth); (b) consumption patterns of the user, such as interests or time of use; and (c) geographic location patterns where the user calls from, receives calls, or is otherwise connected to the network.

The different systems and method disclosed herein aim to capture one or more identifying patterns of the user (such as but not limited to the ones disclosed above), and try to identify the user throughout his usage of the network.

It should be noted that while systems and methods disclosed herein may be used for the identification of multiple network usages by the same anonymous user, the disclosure further covers systems and methods which may be used for identification of multiple network usages by single users which are not entirely anonymous. For example, some users may provide partial details, but such that may not be sufficient for a certain identification of the user over different connection. In another example, a single user may use more that a single identifiable account (e.g. a single user using different e-mail accounts, various cellular phone numbers, etc), in which case the herein disclosed systems and methods may be used for determining that two or more identifiable accounts are used by a single entity. Clearly, the systems and methods disclosed are not limited to those few examples.

Figure 1 illustrate a method for identifying a user which may connect using different details at different times (e.g. a cellular telephony prepaid user), according to an embodiment of the invention. The method may include some or all of the following stages depicted in Fig. 1. First, information may be collected about multiple users (also referred to as customers) of one or more networks. It is noted that users of the multiple users may connect to the network through the different channels of usage.

Next, a fingerprint of the usage of different users may be assembled for future identification (wherein this may be carried out for all users, or only for some subgroups selected according to various selection criteria - e.g. only prepaid users, only users who are power-users, only users that talk for over one hour a day, and so forth).

Then, the stream of usage may be continuously monitored or sampled, in order to identify customers resurfacing with different identities, by comparing information gathered about the resurfacing customer during a later session to finger print information associated with one or more user identities. It is noted that in some cases, where the identification of the user is not clear, additional data may be actively collected to enable better comparison, and/or data may be gathered but not associated to the user identity, until a positive identification is reached (e.g., call history of that call will be logged, but geographical location will only be added to the user identity patterns when the behavioral and call records will not enable identification). The system calculate a similarity number, e.g., ranging from 0 to 1, and repeat a comparison with stored profiles until the similarity number reaches a similarity threshold, in which case, the highest ranking similar profile may be determined to be the user's identity. A unified profile of the customer along time (e.g., including prior numbers or accounts) may be constructed to have accurate statistics.

Based on the customer identity, e.g., the unified profile, a service may be offered or provided to the customer based on information gathered in relation to the customer when the customer connected to the network using different identifying details than those used during to providing of the service. For example, a user who is known to make calls to certain foreign countries may be offered relevant promotions. In another example, promotions may be offered to high- frequency users to induce loyalty and keep them in the network, etc. It is noted that other than providing service to the customer based on such information gathered, other actions may be carried out as well, such as recording user calls, limiting bandwidth allocated to the user, and so forth.

According to an embodiment of the invention, a system that implements the method may have a substantially unique leading identity associated with each user. This unique identifier allows managing all information around a single key. When a new number is detected as belonging to an existing user (identity), it may then be associated with this identity. Thus, each identity may be associated with one or more MSISDNs (mobile phone numbers).

According to an embodiment of the invention, the above disclosed method may be implemented by a system, which includes a monitor for monitoring connections of multiple users of the network, and a processor which is configured to analyze information gathered by the monitor as disclosed in relation to the method, and to identify the recurring patterns. Other details of the method may also be implemented by such a system. Figure 2 illustrates is a block diagram of user identification as may be implemented in systems and methods as herein disclosed, according to an embodiment of the invention. It will be understood that different embodiments of the invention need not have every one of the shown modules. Certain exemplary aspects of the embodiments of the invention shown in Fig. 2 are described below.

The Social Network Analysis (also referred to by the acronym SNA) system builds the implicit social network of the user. For example, using logs from voice and peer to peer services such as SMS, MMS or Instant Messaging for example, the system builds a network representing the social group of the user. Each node in this network represents a user, and nodes are connected by weighted edges if the people the nodes represent are communicating via some mobile channel. The more two people are communicating (like more calls or more SMSs), the weight of the edge connecting them in the network is larger. According to an embodiment of the invention, weights may have decay functions that govern the decreasing of the weight along time automatically according to some function. Thus, occasional, erroneous or limited scope communication between individuals is not captured in the network (and in the fingerprint).

Once the network has been built, hubs, or popular nodes that are common to many people's peer group may be filtered out from the graph. The intention is to have a graph with neighborhoods (the nodes connected to each node representing the people the person is in contact with on a regular basis) as unique to each customer as possible.

Further, according to an embodiment of the invention, if there is a large similarity between nodes (for example, two sisters who have very common communication patterns with their family), the degree of the neighborhood for a specific node is extended. Namely the fingerprint may be representing not just the direct peers, but also the peers of the peers (second degree neighborhood) to find differentiating patterns. The SNA is maintained along time, to capture the evolving social worlds of the customer.

Usage habits fingerprints may be used, according to an embodiment of the invention, in situations where the user cannot be identified based solely on his social fingerprint, even though it may be used in other situations as well. According to an embodiment of the invention, the system will include information about his usage habits to try and make his signature unique. Usage habits patterns may be collected, according to an embodiment of the invention, from value added services (VAS services), and include statistics on how the user is using VAS services such as MMS, SMS, WAP, Mobile Internet etc.

Examples for such statistics for a cellular network scenario may include: (a) consumption distribution of different services (percentage of SMS in the VAS usage, of MMS, etc.); (b) time and week day usage distribution; (c) web content types (interests as hinted by the web pages the user is browsing); (d) time to reach percentage of consumption (namely, how much time it takes the user to consume 80% of his balance); (e) pre-paid indicators such as time interval between top-ups, sum of typical top-up, etc.

According to some embodiments of the invention, the usage habits may be a principal indicator, or they may be used secondarily to supplement or confirm one ore more other principal indicators, e.g., to confirm or deny an identification based on the social network.

When a user makes a call, the BTS (the station that the handset communicated with) is logged. Thus, along time, the system can capture a distribution of where a person usually makes his calls form (ride home, office, house, gym etc.). This geographical pattern is used to further differentiate between several otherwise common identities.

According to an embodiment of the invention, if the SNA and Usage Patterns steps cannot uniquely identify the user, the Geographic Information from the voice service may be used. However, this is not necessarily so; in other embodiments, the geographic usage may be a principal indicator or one of a number of principal indicators.

According to an embodiment of the invention, coactivity may be used an additional measure to differentiate between identities. That is, if two nodes are considered to be the same person based on all other facets, then having those two nodes active in the same time may hint that they are two separate identities and the two will be maintain in the system as such. This is mainly important at the initial phase of a new identity which for a limited period of time resembles an existing identity. Coactivity may enable managing them as two separate identities earlier.

Figure 3 illustrates a state machine which may be implemented either by a system or as a method, for identifying users of a network, according to an embodiment of the invention.

When a number (or other user identifier) is used for connecting of a user to network resources, the system tries to see if it is a new number (e.g. MSISDN). If it is identified as a known active number, it is going through a filtering phase to decide if it belongs to a black list. A black list may include known identities that have been deemed as inappropriate for long term relationships (tourists, foreign workers, people with high complaint rate with customer care, people with poor credit risks, other forms of financial or operational risk, etc.). The system may permit an administrator (or administrator software) to browse this list from time to time and reconsider the status of selected blocked numbers (for example, if marketing policy has changed).

If the number is not an active one, it is moved for the identification phase. Since it is new in the system, one can assume that not enough information is available to construct a fingerprint and make a decision about its identity. Thus, the number stays in the accumulation state until the user has made enough usage of the network so a pattern is emerging. From time to time, the system tries to use the collected information to build a fingerprint that will identify the user. Once it succeeds, the system first checks the black list to see if the identity is not blocked (in case the leading identity the new number has been associated with), and if applicable, manages the number as an identified identity.

Conveniently, from now on, all usage patterns and statistics of this number may be associated with the detected identity.

Figure 4 illustrates a system for user identification, according to an embodiment of the invention. The proposed system may include some or all of the following main components (as well as possibly additional components).

The Social Network Analysis (SNA) engine is responsible for building and maintaining the social fingerprint of the user.

The Customer Experience Management (CEM) engine is responsible for capturing the usage patterns of the customer while using services.

The Location Engine is responsible for building the geographic usage distribution of the calls the user is making.

The Identity Engine manages the identity of customers. It has association between a main identity (e.g. general unique code or MSISDN) and all the MSISDN' s associated with this specific user. The identity engine gets its information from the SNA, CEM and Location engines.

The Customer Browser is an application that allows the operator to browse the identities in the identify engine data base. Using query and filtering the user can digest a specific customer population viable for a promotion.

The systems above are depicted in Figure 4 within the general topology of the consumed data sources and client systems of the operator. According to various embodiments of the invention, on the operator side the system can connect and exchange information with some or all of the following systems:

Pre-Paid management - this would be necessary to get pre-paid usage information (top-up date and time, amount etc), and to update MSISDN profiles (for example, if some benefit is to be given to the customer).

Marketing automation - this would be necessary to forward the marketing automation systems information on groups of customers that will enjoy a campaign.

Billing systems - part of the benefits of the system is to make focused campaigns, thus integration with the billing system to update prices and rates is necessary. Further, the billing system may act as a source for the CEM system to build the profile of the user.

According to different embodiments of the invention, the herein disclosed methods and systems may supports one or both of the followings modes of operation - batch and real time.

Fig. 5 depicts an example use case of batch mode. In batch mode, the system allows the operator to generate timely identification lists of Identities based on specific attributes. For example, the operator can ask the identity server to provide all customers with specific attributes like average ARPU or activity in the network for some consecutive period of time. Once this list of customers has been generated, the user send this list to other systems (marketing automation, billing etc) and the action and management of interaction with the customer is beyond the scope, or involvement of the proposed system.

For example, the operator would access the customer browser and ask for all customers active for the last 6 months, with an average top up sum of $60 and more than 8 top up events that are using voice, SMS and mobile internet. Let this list be deemed List A. The operator marketer then scans every day the list of MSISDNs that have been top-upped in that day and looks for MSISDNs that appear in List A. All customers from List A appearing for the first time after a top-up event are being sent an SMS notifying them that they got 20 minutes of air time for free. As can be seen in the scenario, the proposed system is not part of identifying when a customer is active after a top-up, or delegating the marketing campaign to him.

The batch scenario has the benefits of simple and cheaper integration in comparison with the real time version.

Fig. 6 depicts an exemplary use case for real time mode. In the real time mode, systems within the operator can access the identity server at any time to gain insight about a specific entity identity/identifier (e.g. MSISDN). Thus, the proposed system becomes part of the promotion life cycle and is available to it at any time.

For example, a marketer can access the marketing automation system and select a group of customers. This marketing automation is integrated with the proposed system, so all customer data is up to date. The marketer defines a campaign, and the list of the target customers is maintained in the marketing automation system. When the marketing automation system would like to issue the daily promotion, it can access the identity system in real time to see if more customers have been identified as viable for the campaign. Further, when identifying a specific MSISDN it can approach the identity server for mapping for the main identity.

As can be seen in the diagram above, according to an embodiment of the invention, in the real time scenario, the operator systems work with the proposed system in iterative mode in contrast to the batch scenario where beyond a one-time generation of customer identities, the proposed system was not part of the promotion life cycle.

It should be noted that all the methods disclosed above may be implemented as one or more computer readable media, having computer readable code embodied therein for user identification, the computer readable code including instructions for one or more of the stages of at least one of the above disclosed methods. When those instructions are executed by one or more processors, the aforementioned stages are carried out.

Different systems as disclosed above may include such computer readable media, and/or may be configured to read such media to acquire the instructions, and to execute the read instructions by one or more processors of the system.

While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Claims

CLAIMS What is claimed is:

1. A method of determining an identity of a user of pre-paid mobile communication services comprising:

providing pre-paid mobile communication services to the user;

compiling a usage profile of mobile communication services consumed by said user; comparing said user's usage profile against a database of previously compiled usage profiles;

based on said comparison, determining a similarity of said user's usage profile to a previously compiled usage profile in said database; and

providing the user with a service based on said determined previously compiled usage profile.

2. The method of claim 1 , wherein compiling a usage profile of mobile communication services consumed by said user comprises compiling a social network profile of said user based at least on: identity of parties initiating communication with said user; and identity of parties with whom the user initiates communication.

3. The method according to claims 1 to 2, wherein compiling a usage profile of mobile communication services consumed by said user comprises compiling a usage habits profile of said user based on at least one of the following sets of parameters: type and/or distribution of consumption of communication services by said user; amount of communication services consumed by said user; time of pre-paid services payments; amount of pre-paid services payments; time to reach a predetermined communication services consumption level; and content of network data services consumed by said user.

4. The method according to claims 1 to 3, wherein compiling a usage profile of mobile communication services consumed by said user comprises compiling a geographic profile of said user based on detected geographic locations of said user while communicating over the communication network.

5. The method according to claims 1 to 4, further comprising confirming or rejecting determination of a similarity of a user profile with a previously compiled usage profile based on simultaneous network coactivity of said user and a user associated with said previously compiled usage profile.

6. The method of claims 1 to 5, wherein determining a similarity of said user's usage profile to a previously compiled usage profile in said database comprises:

comparing a first set of parameters of said user's usage profile against said database; determining whether a similarity threshold parameter with a previously stored user profile has been detected; and

if not, repeating for a next set of parameters until said similarity threshold parameter has been reached.

7. The method of claims 1 to 6, wherein said service is selected from the group of services consisting of: a promotion, a customized experience, an identity theft detection.