Trend or pattern trigger device
The present invention implements, in a trend or pattern trigger device or system, the generation of hypotheses automatically while considering any collection of data by automatically assessing with templates or charts the presence of patterns in the data without a priori postulation of what the pattern might be. The trend or pattern examination templates or charts contain quantifiable thresholds for data or change in data so that trends or patterns in data can be discerned inductively and automatically without preconceived notions of what or where the patterns or trends might be.
1 . A method for monitoring and event-triggering reporting regarding a dynamic system containing aggregated data, wherein said aggregated data are analyzed without a priori postulation as to which data, thresholds, trends or patterns will prove important, comprising: a) randomly generating i) at least two charts; or ii) at least two thresholds pertaining to a single chart, wherein each of said thresholds and said charts is a collection of one or more data field definitions in which a datum from a database is defined as to registration (presence or absence) of nominal data or as to threshold quantity or rate of change of quantifiable data without requiring any specific data field definition; as to at least two real-world events in a dynamic system, completely without postulating as to significance of said charts or said thresholds and chart prior to their random generation; b) providing input including data of said real-world events, together with said charts or said thresholds and chart, to analytical software capable of describing said input in dataset reflecting said at least two events without a priori postulation as to which data, thresholds, trends or patterns will prove important; c) calculating, using said analytical software, any coordination between said at least two events and said charts or said thresholds and chart, to identify, without a priori postulation of what any pattern present might be, whether a trend or a pattern is present relevant to said charts or said thresholds and chart; and d) providing an output to a user in a precipitated, triggering way regarding any identification of said trend or said pattern if present to create either a real world transformational triggering event or the absence of a real world transformational triggering event when no trend or pattern has been identified.
2 . The method for monitoring according to claim 1 wherein said chart includes at least three parameters measured over at least four consecutive repeating time units.
3 . The method for monitoring according to claim 1 wherein said chart includes at least four parameters measured over at least five consecutive repeating time units.
4 . The method for monitoring according to claim 1 wherein said output to a user includes an alarm or event which invites the participation of a human intellect.
5 . A method of identifying, by random hypothesis generation without a priori postulation as to which data, thresholds, trends or patterns will prove important, whether a trend or pattern is present in data and creating a triggering event when so, consisting essentially of:
a) designing a database comprising any collection of data pertaining to any system, wherein said data is neither limited to time lapse nor non-time lapse data, and generating a hypothesis describing a pattern or trend if present, further randomly generating said hypothesis by randomly generating i) at least two charts; or ii) at least two thresholds pertaining to a single chart, wherein each of said thresholds and said charts is a collection of one or more data field definitions in which a datum from said database is defined selected from the group consisting of the value of nominal data, threshold quantity of quantifiable data, and rate of change of quantifiable data as to at least two real-world events without a priori postulation as to which data, thresholds, trends or patterns will prove important;
b) providing both said database and said charts or said thresholds and chart to analytical software;
c) calculating, using said analytical software, whether said generated hypothesis has identified any actual trend or any actual pattern in said data; and
d) providing an output to a user in a precipitated, triggering way regarding any identification of a trend or pattern if present to create either a real world transformational triggering event or an absence of any real world transformational triggering event when no trend or pattern has been identified.