Amplero, Inc. - [2018] APO 59

IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Amplero, Inc. [2018] APO 59

Patent Application: 2014392681

Title:Automated marketing offer decisioning

Patent Applicant:                   Amplero, Inc.
Delegate:  Isaac Tan
Decision Date:  5 September 2018
Hearing Date:  12 July 2018 by written submissions

Catchwords: PATENTS – section 45 – examiner objection – whether the invention is a manner of manufacture – substance of the invention lies in the creation and training of a ‘tree’ to send messages to telecom subscriber users to maximise the information gained for an ‘attribute’ – gain or outcome for users is by measuring the amount of revenue generated by one customer – claimed invention is not a manner of manufacture – no technical effect – application refused

Representation: David Webber, Patent Attorney of Davies Collison Cave Pty Ltd

IP AUSTRALIA
AUSTRALIAN PATENT OFFICE
Patent Application: 2014392681

Title:Automated marketing offer decisioning

Patent Applicant: Amplero, Inc.
Date of Decision: 5 September 2018
DECISION
I find that the claimed invention is not for a manner of manufacture.
I direct that the application be refused.
REASONS FOR DECISION
Background

Amplero, Inc. (applicant) filed patent application 2014392681 (application) on 30 April 2014 as international application PCT/US2014/036182 filed under the Patent Cooperation Treaty. The application has an international filing date of 30 April 2014 and claiming a priority date of 29 April 2014.
The Applicant requested examination on 28 October 2016, and has been subjected to three (3) examination reports respectively dated 5 May 2017, 26 April 2018 and 4 May 2018. On 7 May 2018, the Applicant requested to be heard. The 7 May 2018 is also the final date of acceptance, as it is the first business day following the one year anniversary of when the first examination report was issued.
In response to the request to be heard, the Applicant was invited to provide written submissions. The Applicant’s written submission was filed on 12 July 2018.
Specification
The specification states that the invention relates generally to deciding marketing messages having offers to send to a particular customer (user) and, more particularly, but not exclusively to training a tree with branch splits being identified based on maximizing an information gain for a message/user attribute, where each node within the tree includes target and control distributions for a feature measure, the trained tree then being traversed for multiple potential message/user combinations, drawing randomly from feature measure distributions in the tree to determine which user/message combinations to send.
The specification goes on to state that:[1]
[t]he dynamics in today’s telecommunications market are placing more pressure than ever on networked services providers to find new ways to compete. With high penetration rates and many services nearing commoditization, many companies have recognized that it is more important than ever to find new ways to bring the full and unique value of the network to their customers. In particular, these companies are seeking new solutions to help them more effectively up-sell and/or cross-sell their products, services, content, and applications, successfully launch new products, and create long-term value in new business models.
[1] Specification as filed, Page 1 Lines 16 - 22
Traditionally, one approach for marketing a particular product or service is by broadcasting, en masse, generic marketing information to all their customers. However, this approach is said to significantly reduce the likelihood that a customer will take up the offer. Another approach is to perform various types of analysis on customer data in order to better understand a customer’s needs. However, in these instances, the offering is provided to the customer long after the offering is no longer relevant.
To address this, the specification describes a method of automatically training a tree usable to identify marketing offers to send to a particular customer. A tree, in the present context, refers to an undirected graph in which any two vertices are connected by one simple path. For example, in one embodiment, a tree may be a binary tree, a ternary tree, or the like. Similarly, ‘the term “node” may also refer to a leaf, where a leaf is the special case of a node, have a degree of one’.[2] To train the tree:[3]
[m]essages (that include offers) having a plurality of attributes are sent to a target use group, and feature measure results from the messages on the target user group are used together with feature measure results for a related control user group, to train the tree where branch splits inside the tree are identified based on maximizing an information gain from the feature measure results for a message/user attribute, and each node within the tree includes target and control distributions for the feature measure for the associated attribute.
[2] Ibid, Page 5 Lines 6 - 8
[3] Ibid, Page 6 Lines 1 - 8
In context, some examples of these attributes are:[4]
…message attributes, including, but are not limited to, a message content (e.g., the offer); an urgency of a message; a method in which the message is communicated to a user, such as email, Instant Messaging (IM), voice mail (VM), or the like; a tone of the message; a time of the day, week, month, and/or year, in which the message is sent or for which an offer is intended; or any of a variety of other attributes. User attributes, include, but are not limited to, an user’s age; a geographic location of the user; an income status of the user; a usage plan; a plan identifier (ID); a refresh rate for the plan; a user propensity (e.g., a propensity to perform an action, or so forth) or the like. Attributes may also include or otherwise represent information about user clusters, including recharge (of a mobile device) time series clusters, usage histogram clusters, cluster scoring, or the like. Thus, attributes may include a variety of information about user and/or messages.
[4] Ibid, Page 6 Lines 9 - 21
The tree may be further optimised with a variety of feature measures, including, but not limited to an average revenue per user (ARPU), active base percentage (ABP), average revenue per paying user (ARPPU), average margin per user (AMPU), or the like. Figure 6 shows one embodiment of a flow diagram of a process usable for creating the tree with feature measure distributions usable at run-time. To illustrate, one example of a flow diagram of a process usable for creating the tree is reproduced below:
According to the specification:[5]
[b]riefly, process 600 of FIG. 6 employs an approach sometimes referred to as A/B testing, hypothesis testing, or split testing, in which randomized experiments with two variants, A and B, are performed to determine an impact on some feature measure of a user’s behaviour. As messages and users have a plurality of attributes, a plurality of evaluations are performed based on the sending of the messages to then create a tree of branch splits based on those attributes (message or user) that indicate a greatest information gain.
Briefly, an information gain G_n at any node n of the tree may be defined as a difference between an overall entropy H_n(R) at the node and an entropy conditioned on a candidate attribute A_i at that node H_n(R|A), or:
G_n(A_i) = H_n(R)−H_n(R|A_i),
where n=0,1,2,…N−1; R is the feature measure lift random variable of interest, such as ARPU. A similar formulation holds for the feature measure ABP lift, as discussed later.
The information gain is directed towards measuring how much the overall entropy decreases when it is known that attribute A_i takes on a specific value A_i=a_ij, or is limited to a given range of values, A_i≤a_ij. The information gain therefore measures attribute A_i's contribution to the randomness of the data. If assigning a value or range to A_i decreases the overall entropy the most, then attribute A_i and its split point value a_ij should be selected at a given node of the tree. Process 600 then may be employed to evaluate the information gain G_n for each candidate attribute to determine split value candidates in creating the tree.
[5] Ibid, Page 23 Line 20 – Page 24 Line 10
At any time that a tree is completed, it may be used during run-time process 700 of figure 7 to determine which message or messages to send to a particular user. Figure 7 is reproduced below:
Figure 7 is described in the specification:[6]
At block 704, vectors for marketing messages and user attributes are constructed. In one embodiment, the attributes may be concatenated in a same order as that used for the training vectors. Thus, if a user is eligible for 1000 possible marketing messages, a 1000 marketing message/user attribute vectors may be constructed for that user. Similarly, for each other users, a plurality of marketing message/user attribute vectors are constructed.
[6] Ibid, Page 31 Lines 5 - 9
Moving on to block 706:[7]
Continuing to block 706, then for each attribute vector for each user, the tree with the feature measure of interest is traversed to generate a rank offering of marketing messages for the user. When a tree has been traversed to a node within the tree based on matching of attribute values in a user’s vector with the tree node values. At that node, a random drawing is performed from the target distribution and the control distributions to obtain an expected lift as a difference between the randomly drawn values. This is performed for each marketing message for the user, to generate a listing of sampled expected lifts for each marketing message that the user is eligible. The marketing message may then be rank ordered based on the determined sampled lift values for each marketing message. This block is performed for each user, for each message for that user, to generate rank offerings of marketing messages for each user. By selecting randomly from the target and control distributions it may be possible to generate different rank orderings of marketing messages and thereby enable an exploration and exploitation approach to providing marketing messages, and thereby potentially improve upon the results for the feature measure of interest.
[7] Ibid, Page 31 Line 25 – Page 32 Line 10
It is worth noting that where an attribute is missing, or a threshold value of the marketing message has not been met, the system is readily adaptable and the specification provides a number of ways in which this can be done.
The claims
The specification ends with 20 claims of which claims 1, 9 and 17 are independent. The independent claims are reproduced below:
Claim 1
A network device, comprising:
a transceiver to send and receive data over a network; and
one or more processors that are operative to perform actions, including:
creating and training, until it is complete, a tree for a first time that has multiple branches and multiple nodes and that represents user responses affecting a feature measure, wherein the creating and training includes sending a plurality of training messages to a plurality of telecom subscriber users in a target user group, and includes creating the branches to each maximize an information gain for a message attribute or user attribute with respect to the feature measure based on responses of the plurality of telecom subscriber users to the plurality of training messages, wherein the plurality of training messages have attributes corresponding to being sent at different times and with different types of messages, and wherein each node within the tree is associated with a subset of the plurality of telecom subscriber users and includes a target distribution for the feature measure and for the telecom subscriber users in the associated subset and includes a control distribution for the feature measure and for other users in a control user group;
using the tree for the first time to send, to multiple additional telecom subscriber user that are distinct from the plurality of telecom subscriber users in the target user group and that each has attributes based at least in part in prior activities in using telecom services, a plurality of further messages by, for each of the multiple additional telecom subscriber users, traversing the tree based at least in part on the attributes of the additional telecom subscriber user, generating an ordered ranking for the additional telecom subscriber user of the plurality of further messages based on determining a feature measure lift by selecting values from the target and control distributions in the tree, and using the ordered ranking to select one or more of the plurality of further messages to send to the additional telecom subscriber user;
repeatedly adapting the tree to changes over time by, for each of multiple additional times after the first time and after the using of the tree to send the plurality of further messages, retraining the tree for the additional time to correspond to further user responses to additional interactions with respect to the feature measure, wherein the adapting includes changing the branches and the nodes of the tree; and
after each of one or more of the additional times, using the retrained tree for the additional time to select and send additional messages to the plurality of telecom subscriber users.
Claim 9
A non-transitory computer readable storage device, having computer-executable instructions stored thereon, that in response to execution by a processor unit, cause the processor unit to perform operations, comprising:
creating and training, until it is complete, a tree for a first time that has multiple branches and multiple nodes and that represents user responses affecting a feature measure, wherein the creating and training includes sending a plurality of training messages having a plurality of attributes, to a plurality of users in a target user group and includes creating the branches to each maximize an information gain for a message attribute or user attribute with respect to the feature measure based on responses of the plurality of users to the plurality of training messages, wherein each node within the tree is associated with a subset of the plurality of users and includes a target distribution for the feature measure and for the users in the associated subset and includes a control distribution for the feature measure and for other users in a control user group;
using the tree created and training for the first time to send a plurality of further messages to multiple additional users distinct from the plurality of users in the target user group, the sending including, for each of the multiple additional users, traversing the tree based at least in part on attributes of the additional user, generating an ordered ranking for the additional user of the plurality of further messages based on determining a feature measure lift by performing a comparison between randomly selected values from the target and control distributions in the tree, and using the ordered ranking to select one or more of the plurality of further messages to send to the additional user;
repeatedly adapting the tree to changes over time by, for each of multiple additional times after the first time and after the using of the tree to send the plurality of further messages, retraining the tree for the additional time to correspond to further user responses to additional interactions with respect to the feature measure, wherein the adapting includes changing the branches and the nodes of the tree; and
after each of one or more of the multiple additional times, using the retrained tree for the additional time to select and send additional messages.
Claim 17
A network device, comprising:
a transceiver to send and receive data over a network; and
one or more processors that are operative to perform actions, including:
creating and training, until it is complete, a model for a first time that has multiple groups of users each having common user and message attributes, wherein the creating and training includes sending a plurality of training messages to a plurality of users in a target user group and includes separating the plurality of users into the multiple groups of users to maximize an information gain for a message attribute or user attribute with respect to a feature measure, wherein each group of users has an associated target distribution for the feature measure and for the users in the group and includes a control distribution for the feature measure and for other users in a control user group;
using the model created and trained for the first time to send a plurality of further messages to multiple additional users distinct from the plurality of users in the target user group, the sending including, for each of the multiple additional users, employing the model to generate an ordered ranking for the additional user of the plurality of further messages based on determining a feature measure lift, and using the ordered ranking to select one or more of the plurality of further messages to send to the additional user;
repeatedly adapting the model to changes over time by, for each of multiple additional times after the first time and after the using of the model to send the plurality of further messages, retraining the model for the additional time to correspond to further user responses to additional interactions with respect to the feature measure, wherein the adapting includes modifying at least one of the target distribution or the control distribution for each of one or more of the groups of users of the model; and
after each of one or more of the multiple additional times, using the retrained model for the additional time to select and send messages.
The Examiner Objection
According to Examination report number 3, which was issued on 4 May 2018 (examination report), the examiner raised an objection under the heading ‘Patentable subject matter’ indicating that the claims do not define a manner of manufacture. The relevant parts of this objection states:
The substance of the claimed invention is to be determined by considering the claimed invention’s actual or alleged contribution to the art. With regards to this, the description reads that due to the competitiveness of the telecommunications market, new methods of “up-selling” or “cross-selling” their products or services are needed. It also discusses launching new products and increasing the longevity of the business models of businesses. It further reads that many vendors continue to seek better tactics to marketing their products in order to address that changing needs of the market.
The application tries to address this problem by “automatically training a tree” [see page 4]. In the field computer science a tree structure is basically a data structure that represents hierarchy within a collection of data. Tree structures represent nodes and branches as dynamically allocated records with pointers to their children and their parents. The method of stepping through the items in a tree is known as traversing. One form of a tree is called a decision tree where the tree-like graph is used as a decision support tool (or possible consequences). These are considered as the basic fundamentals of a tree structure or a tree like graph.
According to the description, the purpose of creating a tree is to make automated marketing offer decisions [see page 21]. This is done by creating or training a tree based on a targeted sample size (using feature measure distributions). Thus it uses a variety of user attributes from the messages and user groups to create or train a tree. Once a tree root node is initialized it is further “trained” or “re-trained”. This is done so by expanding the branches by adding more message attributes or simply customer data.
Based on your response and with regard to the above, the highlighted arguments simply depict actions that are fundamental to the logical decision making process in the field of computer science. This is especially evident when you remove the context of “telecommunication industry” from the application of the theory of decision trees. Thus the alleged invention is simply using already known theories of computer science to make a marketing decision. In reply to your response that the invention is directed to a specific advance in artificial intelligence technology is not persuasive enough. This is simply a decision making process that uses conditional processes to achieve an outcome.
Therefore, the substance of the present invention relates to a decision making algorithm which which [sic] improves making marketing decisions in telecommunications (while using generic computer implemented decision making techniques).
Applicable Law

The request for examination of the present application was filed on 28 October 2016. As a consequence, substantive amendments to the Patents Act 1990 (the Act) bought about by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 that came into effect on 15 April 2013 apply to the present application.
Therefore, the standard of proof that applies in the present case is the balance of probabilities. I must accept the application if I am satisfied on the balance of probabilities that the application complies with the Act. If I am not so satisfied, then I can refuse the application.
Section 18 of the Act provides for a standard patent:
(1) Subject to subsection (2), an invention is a patentable invention for the purposes of a standard patent if the invention, so far as claimed in any claim:
(a) is a manner of manufacture within the meaning of section 6 of the Statute of Monopolies; and
(b) when compared with the prior art base as it existed before the priority date of that claim:
(i) is novel; and
(ii) involves an inventive step; and
(c) is useful; and
(d) was not secretly used in the patent area before the priority date of that claim by, or on behalf of, or with the authority of, the patentee or nominated person or the patentee's or nominated person's predecessor in title to the invention.
Case Law on Manner of Manufacture
In National Research Development Corporation v Commissioner of Patents [1959] HCA 67 (NRDC), the High Court provided a statement of the law in this regard:[8]
... a process, to fall within the limits of patentability which the context of the Statute of Monopolies has supplied, must be one that offers some advantage which is material, in the sense that the process belongs to a useful art as distinct from a fine art ...- that its value to the country is in the field of economic endeavour.
[8] National Research Development Corporation v Commissioner of Patents [1959] HCA 67; (1959) 102 CLR 252, [22]
In discussing the “vendible product” proposition put forward by Morton J in Re G.E.C’s Application, (1942) 60 RPC 1, the High Court in NRDC upheld the validity of a patent for the use of previously unknown properties of a known chemical to effect a new purpose:[9]
The effect produced by the appellant’s method exhibits the two essential qualities upon which “product” and “vendible” seem designed to insist. It is a “product” because it consists in an artificially created state of affairs, discernible by observing over a period the growth of weeds and crops respectively on sown land on which the method has been put into practice. And the significance of the product is economic; for it provides a remarkable advantage ... for one of the most elemental activities by which man has served his material needs, the cultivation of the soil for the production of its fruits.
[9] Ibid, [25]
The High Court though was not laying down a precise formulation that can be applied unthinkingly. In D’Arcy v Myriad Genetics Inc [2015] HCA 35 (Myriad) the court considered that:[10]
This Court in NRDC did not prescribe a well-defined pathway for the development of the concept of “manner of manufacture” in its application to unimagined technologies with unimagined characteristics and implications. Rather, it authorised a case-by-case methodology.
[10] D'Arcy v Myriad Genetics Inc [2015] HCA 35, [23]
That case-by-case approach must have regard to the substance of the claimed invention, not simply the form of the claim. The point was made succinctly in the Myriad case by Gageler and Nettle JJ:[11]
Whatever words have been used, the matter must be looked at as one of substance and effect must be given to the true nature of the claim.
[11] Ibid, [144]
In Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177 (RPL), the Full Court of the Federal Court stated the same thing in the context of an invention that was in substance a scheme:[12]
A claimed invention must be examined to ascertain whether it is in substance a scheme or plan or whether it can broadly be described as an improvement in computer technology. The basis for the analysis starts with the fact that a business method, or mere scheme, is not, per se, patentable. The fact that it is a scheme or business method does not exclude it from properly being the subject of letters patent, but it must be more than that. There must be more than an abstract idea; it must involve the creation of an artificial state of affairs where the computer is integral to the invention, rather than a mere tool in which the invention is performed. Where the claimed invention is to a computerised business method, the invention must lie in that computerisation. It is not a patentable invention simply to “put” a business method “into” a computer to implement the business method using the computer for its well- known and understood functions.
[12] Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177, [96]
and that:[13]
It is not a question of stating precise guidelines but of deciding, in each case, whether the claimed invention, as a matter of substance not form, is properly the subject of a patent.
[13] Ibid, [98]
In Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150 (Research Affiliates), the Full Court of the Federal Court noted that:[14]
[14] Research Affiliates LLC v Commissioner of Patents [2014] FCAFC 150, [94] citing Grant v Commissioner of Patents [2006] FCAFC 120; (2006) 154 FCR 62; Business Machines Corporation v Commissioner of Patents [1991] FCA 625; (1991) 33 FCR 218; CCOM Pty Limited v Jiejing Pty Limited [1994] FCA 1168; (1994) 51 FCR 260; Welcome Real-Time SA v Catuity Inc [2001] FCA 445; (2001) 113 FCR 110
When the authorities in Australia prior to and including Grant are considered, a consistent approach emerges as to the relevance of:
a distinction between a claim to a business scheme and claims to methods which in practice result in a new machine or process or an old machine giving a new and improved result – that is, a distinction between mere intellectual information and a method that affects the operation of an apparatus in a physical form (Grant at [18]);
the fact that the claimed steps are foreign to the normal use of computers, such as the production of an improved curve image (IBM 2 at 225-226);
the particular mode or manner of achieving an end result which is an artificially created state of affairs, such as the storage of data as to Chinese characters and retrieval of graphic representations to enable word processing (CCOM at 295);
whether part of the invention is an inventive method which includes the application and operation in a physical device (Grant at [30]);
the distinction drawn in Catuity, as explained in Grant (at [24]), between “a technological innovation which is patentable and a business innovation which is not”. In Catuity, Heerey J did not accept that a physically observable effect was necessarily required (at [128]) but the Full Court in Grant expressed the opinion that a physical effect in the sense of a concrete effect or phenomenon, or manifestation or transformation is required (at [32]).
the fact that a physical effect is required does not make it sufficient to confer patentability;
the fact that a method may be called a business method does not prevent it being properly the subject of letters patent (Grant at [26] citing Catuity at [125]-[126]);
the fact that for claimed computer programs, the courts look to the application of the program to produce a practical and useful result, so that more than “intellectual information” is involved (Grant at [29]). A method that is in the nature of directions for use does not constitute an invention or a manner of manufacture in the absence of some previously unrecognised property of an aspect of the method (Grant at [29]).
In determining whether an invention contains patentable subject matter, the delegate in Aristocrat Technologies Australia Pty Limited [2016] APO 49 (Aristocrat) provided the following summarisation:[15]
[15] Aristocrat Technologies Australia Pty Limited [2016] APO 49, [35]
I conclude that it is relevant to consider a range of matters. Without seeking to be exhaustive, these include:
·there must be more than an abstract idea, mere scheme or mere intellectual information;
·is the contribution of the claimed invention technical in nature;
·does the invention solve a technical problem within the computer or outside the computer;
·does the invention result in improvement in the functioning of the computer, irrespective of the data being processed;
·does the application of the method produce a practical and useful result;
·can it be broadly described as an improvement in computer technology;
·does the method merely require generic computer implementation;
·is the computer merely an intermediary or tool for performing the method while adding nothing of substance to the idea;
·is there ingenuity in the way in which the computer is utilised;
·does the invention involve steps that are foreign to the normal use of computers; and
·does the invention lie in the generation, presentation or arrangement of intellectual information.
However, in Todd Martin [2017] APO 33, the delegate made the following comments regarding the “range of matters” set down in Aristocrat:[16]
With this list of points, the delegate in the Aristocrat case was not intending to indicate a list of conditions, for computer-implemented cases, to be met to define a manner of manufacture. That is evident from the delegate’s statement that the list was not intended to be exhaustive. Moreover, it would appear improper to find there was a manner of manufacture simply on the basis that one or more points could be answered in favour. In the present case for example, it may be that at least the fifth dot-point, regarding whether the application of the method produces a practical and useful result, is satisfied. On the other hand, that consideration on its own would be insufficient in the present case. Conversely it would also appear improper to find there was no manner of manufacture simply on the basis that one or more points could not be answered favourably. Rather than these points being seen as conditions to be met, they should be seen as relevant matters to consider, as the delegate stated at [35]. The substance and contribution of the claimed invention in each case should be considered on its merits overall and various points under the law would appear to have varying degrees of relevance depending on the case.
[16] Todd Martin [2017] APO 33, [47]; Similar consideration was also applied in Bio-Rad Laboratories, Inc. [2017] APO 38, [37]
Manner of Manufacture Submissions and Considerations
As set out above, the claimed invention consists of independent claims 1, 9 and 17. Claim 1 and 9 are directed towards an iterative approach of ‘creating and training’ a tree, and is divided in to three stages. Claim 17 is broader, as it refers to a ‘model’ rather than a ‘tree’. However the specification makes it clear that a ‘tree’ is just one embodiment, and in alternative embodiments, may be substituted with a selection of other ‘models’.[17]
[17] Specification as filed, Page 7 Lines 11 - 16
According to the Cambridge dictionary, the term ‘training’ is defined as ‘the process of learning the skills you need to do a particular job or activity.’ Put another way, in the present context, ‘training’ could be said to be the manner in which a logical decision making process is optimised. By way of example, a ‘tree’ could be visualised as a flowchart consisting of a plurality of ‘branches’ and ‘nodes’. Each node represents a ‘message’ or question, and each branch is an answer to the question. A question may also relate to whether a condition is met or an attribute is present. Each branch may also lead to an additional node, which represents a different question.
In the first stage, the ‘training’ involves asking a question to users in a selected group, and creating a plurality of ‘branches’ to account for possible responses which may be received. Each response may be further segmented based on any number of attributes, so that the ‘information gain’ for each question, is maximised. That is, the first stage seeks to identify the subset of users who may respond favourably to particular questions. In the second stage, the tree created in the first stage is used with a different, additional group of users. Depending on the attribute of this additional group of users, relevant questions are selected and ranked, in order to select which question to send to the additional group of users. The third stage involves repeating the first stage and the second stage, so that the tree can be further ‘adapted’ or refined. This is done by changing the ‘branches’ and ‘nodes’ of the tree.
In the examination report, the examiner has characterised the substance of the invention to be ‘a decision making algorithm which which [sic] improves making marketing decisions in telecommunications (while using generic computer implemented decision making techniques).’
The Applicant disagrees, stating that the examiner has oversimplified or mischaracterised the invention. The Applicant did not provide a definite characterisation of what they consider the substance of the invention to be. Rather, the Applicant submits that:
The invention relates to new computer technology for processing message and user attributes in order to create a unique tree or model structure, having branches with nodes that are based on a selected subset of user and message attributes and that include target and control group feature measure distributions for each such node. The structure is used to generate feature lift values based on the distributions in the structure in order to determine messages to be sent to users. The claims also refer to how the structure and the distributions are then continually adapted based on the user responses received.
In my view, it seems clear to me, that the substance of the invention lies in the creation and training of a ‘tree’, that serves as a decision making tool, to send messages to telecom subscriber users to maximise the information gained for an ‘attribute’. The information gain is determined based on a ‘feature measure’.
It is worth noting that the specification places some emphasis on the data used to form the ‘attributes’ and ‘feature measure’. In brief, the term ‘attribute’ can be classified as a message attribute or a user attribute. According to the specification:
…message attributes, including, but are not limited to, a message content (e.g., the offer); an urgency of a message; a method in which the message is communicated to a user, such as email, Instant Messaging (IM), voice mail (VM), or the like; a tone of the message; a time of the day, week, month, and/or year, in which the message is sent or for which an offer is intended; or any of a variety of other attributes. User attributes, include, but are not limited to, an user’s age; a geographic location of the user; an income status of the user; a usage plan; a plan identifier (ID); a refresh rate for the plan; a user propensity (e.g., a propensity to perform an action, or so forth) or the like. Attributes may also include or otherwise represent information about user clusters, including recharge (of a mobile device) time series clusters, usage histogram clusters, cluster scoring, or the like. Thus, attributes may include a variety of information about user and/or messages.
Similarly, the ‘feature measure’ may be based on average revenue per user (ARPU), active base percentage (ABP), average revenue per paying user (ARPPU), average margin per user (AMPU), or the like. It is worth noting that these terms are typically used by companies that offer subscription services, such as telecommunication providers, as a measure of the revenue generated by one customer. At this point, it is clear that the premise of the invention resides in a business method or scheme. However, RPL makes it clear that[18] ‘[t]he fact that it is a scheme or business method does not exclude it from properly being the subject of letters patent, but it must be more than that.’
[18] Commissioner of Patents v RPL Central Pty Ltd [2015] FCAFC 177, [96]
In relation to the creation and training of a ‘tree’, the Applicant provides the following submissions:
As recited in the claims, the tree or structure is generated uniquely by selecting, from numerous possible attributes of users and messages, a subset of the user and message attributes to represent within the tree. In particular, since it is computationally impossible to consider all combinations of attributes as the number of attributes grows large, such a new type of tree structure allows the most important attributes to be concisely represented in a form that is useful. The tree begins with a root node that represents all user attributes and all message attributes, such that all user and message pairs match the root node. Each new branch of the tree (and its corresponding new node) is then created in a way to maximize the efficiency of the tree structure, based on selecting a particular user or message attribute that provides the most statistical information gain relative to the next-higher-level node in the tree with respect to a criteria being measured. In this manner, the new node for that new branch represents a subset of the total possible users and messages who have attributes that match those of the new node and its parents in the tree. By recursively creating the tree in this manner, leaf nodes at the bottom of the tree are quickly reached so that each represents a different combinations of users and messages. Once the tree is created, its structure can then be used to quickly assess how useful a particular new user and message pair combination is, by traversing the tree for such a user and message pair to find the lowest node in the tree whose associated attributes match that user and message for the pair, and then using both the target and control distributions for that matching node to determine the usefulness of that user and message pair to other users and relative to other user and message pairs.
The particular features recited in the claims provide technical advantages that allow large amounts of data to be processed in ways that the prior art cannot. In addition, the creation and use of, for each node in the tree, both a target distribution and a control distribution for the selected message/user attribute represented by that node allows a more precise determination of the differential effects of those message/user attributes than would a comparison to, for example, a single control group for the whole tree. Thus, these techniques allow types of processing to be performed that could not otherwise be done.
It appears that the Applicant’s position is that the claimed invention provides a technical solution for adapting to, and processing large amounts of information in a manner that improves the gain or outcome for users who receive messages. Generally, if it can be shown that the claimed invention ‘solves a technical problem relating to the running of computers generally’[19], then such an invention may constitute patentable subject matter. On this point, I note that the specification states:[20]
It is further noted that while a tree structure is described herein one embodiment usable to maximize an information gain for a message or user attribute, other models may also be used. Thus, other embodiments of the innovations disclosed herein may include other models including, but not limited to logistic regression models, neural networks, support vector machine regression models, Gaussian Process models, General Bayesian model, and so forth.
[19] Apple Inc. [2018] APO 54, [69], [71]
[20] Specification as filed, Page 7 Lines 11 - 16
It is also worth noting that the information processed by the claimed invention, is on the basis of a business or commercial decision, as opposed to one of a technical nature. Specifically, the improvement in the gain or outcome for users who receive messages is determined by measuring the amount of revenue generated by one customer.

In this respect, the substance of the invention relates to a scheme of implementing a marketing strategy. That is, the invention has taken the idea of performing various types of analysis on customer data, and has utilised a computer system so as to perform this in a quicker, more efficient manner. On this point, the Applicant submits that ‘[t]here is nothing conventional about the tree structure and nodes that are created and adapted as described in the specification and claims, and in particular the use of the target group and control group distributions in each node.’ This may be so, but what is unconventional resides in the nature of data itself, not in the tree structure.
While I acknowledge that there is a degree of complexity involved in this analysis, as shown by the numerous examples described in the specification, the level of detail with respect to the computer implementation makes it clear that the computer is simply used as a tool. In a general sense, a computer system would typically be used to run software, comprising a set of rules or problems for the computer so solve. The software and rules would be unique, depending on the intended specific application of the computer system. The use of ‘trees’, or any of the models briefly mentioned in the specification[21], are tools of logical and mathematical decision making. A mere application in business is not an advance in technology.
[21] Ibid, Page 7 Lines 11 - 16
In Trading Technologies International, Inc [2017] APO 13, it was submitted that programming a computer results in improving the computer by granting it functionality it previously did not have. However I disagreed with this line of reasoning stating that:[22]
The difficulty I have with this statement is that if I proceed on the basis that a computer system is automatically patentable if provided with functionality it previously did not have, then this would thereby extend to any computer system that is programmed in some manner regardless of how rudimentary said functionality may be. In order to utilise a computer system, said system would need to be initially programmed with software.
[22] Trading Technologies International, Inc [2017] APO 13, [31]
Lastly, the Applicant submits that:
It needs to be borne in mind it is computationally impossible to individually consider every combination of attributes to determine and rank the best combinations, and this type of structure, with the target and control feature measure distributions, allows a subset of the most important attributes (with respect to the criteria being measured and used to create the structure) to be identified and used in an efficient manner, based on measuring and using the attributes with the greatest information gain.
Generally, I agree with the Applicant’s statement. However, this is not the test. The question which should be asked is whether, in running the scheme, the computer is carrying out anything other than the usual functions of a computer. On this point, I am inclined to say no. If this was the case, then I would have expected the specification to have included information that would allow the person skilled in the art to implement the invention on a computer that did not have the capability, in its normal operation, to do so. Such information is not present.
Rather, the specification simply provides, in the most general of terms, an overview on the technical aspects of the invention. In relation to how the ‘tree’ is generated and retrained, the information provided is all of a business or commercial nature. Similarly, the measurement of success is not one of a technical improvement. Instead, it in measuring the amount of revenue generated by one customer. At its very essence, the substance of the invention relates to a business scheme.
Consequently, I am not satisfied that on balance, the present invention as claimed would qualify as being a manner of manufacture. This applies to all of the claims and I see no patentable subject matter within the present application that could be made the subject of a claim so as to result in that claim being for a manner or manufacture.
Conclusion
In this case, I have found the claimed invention is not for a manner of manufacture. Moreover, I consider that there is nothing in the specification that could overcome the finding of lack of manner of manufacture.
In these circumstances, it is appropriate that the application be refused.
Isaac Tan
Delegate of the Commissioner of Patents