WE CLAIM:
1. A time-series data processing device comprising:
an event waveform extracting unit for extracting waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
a co-occurrence rate calculating unit for calculating a co-occurrence rate of the waveform data extracted by the event waveform extracting unit among time-series data;
a grouping unit for classifying the time-series data into groups depending on the co-occurrence rate of the waveform data calculated by the co-occurrence rate calculating unit; and
an event information generating unit for determining time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified by the grouping unit, and generating event information identifying an event related to the waveform data on a basis of the determined time.
2. The time-series data processing device according to claim 1, further
comprising:
a detecting unit for detecting outlier data from the time-series data, the outlier data being partial string data falling within an abnormality range; and
a determining unit for determining abnormality of the subject equipment on a basis of the outlier data detected by the detecting unit and the event information generated by the event information generating unit.

3. The time-series data processing device according to claim 1,
wherein the event waveform extracting unit extracts the waveform data on a basis of a combination of partial string data whose values continuously increase or decrease in the time-series data.
4. The time-series data processing device according to claim 1,
wherein the event information generating unit generates the event information on a basis of a starting time and an ending time of the waveform data, a duration of the waveform data, descriptive statistics of the waveform data, a maximum amplitude and a frequency of the waveform data, and a type of the waveform data.
5. The time-series data processing device according to claim 1,
wherein the co-occurrence rate calculating unit calculates any of a first co-occurrence rate, a second co-occurrence rate, and a third co-occurrence rate,
the first co-occurrence rate being a value obtained by dividing a number of times at which periods of occurrence of the waveform data overlap with each other among the time-series data by a number of the waveform data occurring in one of the time-series data,
the second co-occurrence rate being a value obtained by dividing the number of times at which the periods of occurrence of the waveform data overlap with each other among the time-series data by a number of the waveform data occurring in another of the time-series data, and
the third co-occurrence rate being a harmonic mean of the first co-occurrence rate and the second co-occurrence rate.

6. The time-series data processing device according to claim 1,
wherein the co-occurrence rate calculating unit generates event time-series data representing a number of occurrences of the waveform data at each time in the time-series data, determines a candidate for a period of occurrence of an event on a basis of a number of occurrences of an event waveform represented by the event time-series data, and calculates, as the co-occurrence rate of the waveform data, numerical data representing whether or not the waveform data occur during the determined candidate period.
7. The time-series data processing device according to claim 1,
wherein the event information generating unit calculates a histogram on time of occurrence of the waveform data in the time-series data, and estimates time of a bin with a maximum frequency in the histogram as time of occurrence of an event related to the waveform data.
8. The time-series data processing device according to claim 1,
wherein the event information generating unit generates event information including a starting time and an ending time of an event, or generates event information including, in addition to the starting time and the ending time of the event, at least one of a duration of the waveform data, descriptive statistics of the waveform data, a maximum amplitude and a frequency of the waveform data, a type of the waveform data, and a band model.
9. The time-series data processing device according to claim 1, further
comprising:

a presentation unit for presenting information on the waveform data included in each of a plurality of groups into which the time-series data are classified by the grouping unit;
an operation inputting unit for receiving an input of operation on presentation of the presentation unit; and
an editing unit for editing the information presented by the presentation unit on a basis of the input of operation received by the operation inputting unit, and outputting a result of edition to the event information generating unit.
10. The time-series data processing device according to claim 9,
wherein the presentation unit presents at least one of a list of groups, a list of the time-series data included in a group, a graph of the waveform data, the waveform data whose periods of occurrence overlap with each other among the time-series data included in a group, a list of the waveform data included in the time-series data, a histogram on time of occurrence of the waveform data, and a band model.
11. A time-series data processing device comprising:
a detecting unit for detecting outlier data from each of a plurality of time-series data sequentially observed over time from subject equipment, the outlier data being partial string data falling within an abnormality range;
a co-occurrence rate calculating unit for calculating a co-occurrence rate of the outlier data detected by the detecting unit among time-series data;
a grouping unit for classifying the time-series data into groups depending on the co-occurrence rate of the outlier data calculated by the co-occurrence rate calculating unit; and

an event information generating unit for determining time at which periods of occurrence of the outlier data overlap with each other among the time-series data included in a group into which the time-series data are classified by the grouping unit, and generating event information identifying an event related to the outlier data on a basis of the determined time.
12. A time-series data processing system comprising:
an event waveform extracting unit for extracting waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
a co-occurrence rate calculating unit for calculating a co-occurrence rate of the waveform data extracted by the event waveform extracting unit among the time-series data;
a grouping unit for classifying the time-series data into groups depending on the co-occurrence rate of the waveform data calculated by the co-occurrence rate calculating unit;
an event information generating unit for determining time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified by the grouping unit, and generating event information identifying an event related to the waveform data on a basis of the determined time;
a detecting unit for detecting outlier data from the time-series data, the outlier data being partial string data falling within an abnormality range; and
a determining unit for determining abnormality of the subject equipment on a

basis of the outlier data detected by the detecting unit and the event information generated by the event information generating unit.
13. A time-series data processing method comprising:
a step of extracting, by an event waveform extracting unit, waveform data estimated to be changed because of an event having occurred in subject equipment from each of a plurality of time-series data sequentially observed over time from the subject equipment;
a step of calculating, by a co-occurrence rate calculating unit, a co-occurrence rate of the waveform data extracted by the event waveform extracting unit among the time-series data;
a step of classifying, by a grouping unit, the time-series data into groups depending on the co-occurrence rate of the waveform data calculated by the co-occurrence rate calculating unit; and
a step of determining, by an event information generating unit, time at which periods of occurrence of the waveform data overlap with each other among the time-series data included in a group into which the time-series data are classified by the grouping unit, and generating event information identifying an event related to the waveform data on a basis of the determined time.
14. The time-series data processing method according to claim 13, further
comprising:
a step of detecting, by a detecting unit, outlier data from the time-series data, the outlier data being partial string data falling within an abnormality range; and a step of determining, by a determining unit, abnormality of the subject

equipment on a basis of the outlier data detected by the detecting unit and the event information generated by the event information generating unit.