Category: G06F17/60

Spam classification/MICROSOFT – T 0022/ 12 – 16 November 2015

This decision concerns the classification of emails where a user flags spam in addition to a computer-implemented method. This is considered to be a de-automation of a computer-implemented method. De-automation is not, according to the Board, a technical solution to a technical problem.

Object of the Invention:

  • the invention concerns the classification of emails, e.g. as either spam or legitimate mail
  • an incoming email is first analysed to determine whether it contains one or more features in a set of predetermined features that are particularly characteristic of spam
  • two types of feature are used: word-oriented and “handcrafted“
  • the former refers to the presence of particular words, or stems of words, the latter to features determined through human judgement alone
  • examples of a handcrafted feature is a sender address, since most spam messages are sent at night from “.com” or “.net” domains

Board I (inventive step):

  • the Board agrees with the Examining Division that the classification of messages as a function of their content is not technical per se
  • it is immaterial whether the messages are electronic messages, because, even though an email has technical properties, it is the content of the email that is classified
  • mathematical methods as such are not technical and the application of a mathematical method as such in a non-technical analysis of message content does not change that
  • if there is a technical effect, it can only reside in the automation of the email classification using a computer
  • the technicality of the computer is not enough to establish a technical effect of any method that it executes

Appellant I (inventive step):

  • a classification based on a combination of “handcrafted features” and “word oriented features” had the technical effect of reducing processing load

Board II (inventive step):

  • the Board is not persuaded that the alleged effect is actually achieved by the invention
  • there is no link between the word-oriented and handcrafted features, so that the latter reduces the processing involved in the former
  • the handcrafted features are, rather, a different class of features that the user considers indicative of spam, but which cannot be expressed in terms of the presence of individual words
  • simply adding a second class of features to the analysis increases the load rather than reducing it
  • furthermore, the Board does not consider that the de-automation of a computer-implemented method, by making a human perform steps that a computer could do automatically, is a technical solution to a technical problem
  • any reduction in computer processing would be a mere consequence of the de-automation
  • handcrafted features relate to information content that is considered as indicative of spam
  • including such features in the analysis might, if well chosen, improve the quality of the classification, but the designation of a second class of features does not provide a technical effect

Appellant II (inventive step):

  • the appellant argued, however, that there was a technical effect in the particular combination of an SVM and a sigmoid function
  • performing the method in two stages, first using an SVM, and then applying an adjustable sigmoid function as a threshold to the output of the SVM, reduced the processing load, which reduced the complexity of the computer implementation
  • thus, the invention was motivated by technical considerations of the computer implementation

Board III (inventive step):

  • the Board is not persuaded by the appellant’s arguments on this point
  • the Board does not find support, anywhere in the application, for the classifier being updated by adjusting the sigmoid parameters alone, without retraining the SVM
  • the generation of parameters for the classifier during the training phase involves two steps:
    • first the weight vector w is determined by conventional SVM training methods
    • second the optimal sigmoid parameters are calculated by using a maximum likelihood on the training data
  • there is nothing to suggest that re-training may involve only one of those steps, or that the classifier may be updated by simply adjusting the parameters A and B
  • on the contrary, it is the teaching of the application that, when the conditions of what is considered as spam change (e.g. when the user reclassifies a message) the whole classifier is retrained
  • furthermore, the Board does not consider that reducing the complexity of an algorithm is necessarily a technical effect, or evidence of underlying technical considerations
  • that is because complexity is an inherent property of the algorithm as such
  • if the design of the algorithm were motivated by a problem related to the internal workings of the computer, e.g. if it were adapted to a particular computer architecture, it could, arguably, be considered as technical (T 1358/09 referring to T 258/03)
  • however, the Board does not see any such motivations in the present case
  • thus, the Board is not persuaded that the use of an SVM in combination with a sigmoid threshold function contributes, technically, to the the computer implementation
  • the Board rather considers this to be a mathematical method
  • the technical implementation of the method consists in programming the computer to perform the method steps
  • this would have been a routine task for the skilled programmer
  • –> no inventive step

Stock index/NASDAQ – T 1161/04 – 6 December 2006

In this case, the responsible board decides in particular whether input data is functional data / a physical entity or simply cognitive data.

Object of the Invention:

  • Claim 1 relates to an apparatus for rebalancing a stock index
  • stock indexes are used to track the performance of a group of stocks
  • capitalization-weighted indexes are regarded as having the disadvantage that a few large stocks may dominate the overall performance of the index
  • the invention is aimed at overcoming this drawback by scaling down large individual stocks and distributing the corresponding excess capitalization over the smaller stocks
  • the output data of the apparatus correspond to the redistributed capitalization weights of the stock index.

Board I (inventive step):

  • the hardware defined in claim 1 is well known as such
  • the invention concerns the way data relating to stocks in a stock index are processed
  • the computer implementation of this process must be assumed to be straightforward

Appellant I (inventive step):

  • decisions T 115/85 and T 362/90 established the principle that automatic visual display of conditions prevailing or desirable in an apparatus or system was basically a technical problem

Board II (inventive step):

  • the invention in the case T 115/85 was a method for displaying one of a set of predetermined messages
  • each such message indicated a specific event which might occur in the input/output device of a text processing system
  • this system comprised a processor, a keyboard, a display and a memory and would also require means for detecting events
  • deciding board: “giving visual indications automatically about conditions prevailing in an apparatus or system is basically a technical problem”
  • the Board cannot agree with the appellants that the present invention relates to an “apparatus or system” in the way these terms are used in decision T 115/85
  • the data entering the system of claim 1 are “information relating to stocks”
  • this is information of a descriptive kind having exclusively “cognitive content” in the sense of decision T 1194/97
  • applying the test proposed in this decision for cognitive information as opposed to “functional data“, it can be seen that if stock information were lost the claimed apparatus would still function, ie still perform the algorithm and produce output data (although these data would be meaningless)
  • presentations of information having merely “cognitive content” are as such excluded from patentability under Article 52(2) EPC and cannot contribute to an inventive step
  • it appears necessary to point out that the nature of such information is irrelevant
  • a description of a gear box may intuitively appear more “technical” than a play by Shakespeare, but in fact both are examples of data having only cognitive content
  • the Board does not accept the appellants’ argument that the present invention provides visual indications automatically about conditions prevailing in a system in the sense of decision T 115/85 for the triple reason:
    • that the conditions are outside the claimed system,
    • that they are not detected by the claimed system but input to it in the form of descriptive data, and
    • that the system in which the conditions prevail is of a commercial rather than technical nature

Appellant II (inventive step):

  • the “information relating to stocks” in claim 1 represents physical entities and therefore has technical character.

Board III (inventive step):

  • in T 208/84 a distinction is made between abstract concepts on the one hand and technical processes involving and modifying a “physical entity”, such as an electrical signal, on the other hand
  • the present invention is not completely abstract since it involves electrical signals
  • but the mere fact that an invention involves signals representing data does not necessarily imply that it solves a technical problem going beyond that of physically representing these data
  • for example, information having only cognitive content is also conventionally represented by (electrical) signals
  • decision T 208/84 therefore requires that the data should represent not just numbers but a “physical entity“
  • the physical entity in that case was an image, and the invention aimed at restoring it if distorted
  • in the case T 1194/97 the physical entity was a synchronization signal
  • the data were thus in both cases “functional” because the degree of restoration of an image or the synchronisation state of a receiver are objectively measurable entities
  • in both cases a technical effect was achieved by the functional nature of the data irrespective of their cognitive contents
  • in the present case however the data represent nothing but numbers, arguably describing “physical entities“, which necessarily require interpretation by a human being without any further interaction with, or modification of, the technical system
  • the data input to the claimed apparatus have no technical function
  • the processing performed on them comprises classification, scaling and redistribution
  • these steps concern exclusively the cognitive content of the data (their numerical value)
  • this is pure information processing which is as such excluded as a mental act by virtue of Article 52(2) EPC
  • the technical task is reduced to the implementation of the process on a conventional computer, something which was obvious for the skilled person –> no inventive step