Processing method and device for data of well site test based on knowledge graph
The present invention provides a processing method and device for data of a well site test based on a knowledge graph. The processing method for the data of the well site test based on the knowledge graph comprises: carrying out format identification on received historical data of the well site test to generate format identification results; establishing a mind map according to the format identification results; generating the knowledge graph of the data of the well site test according to the mind map; and processing the historical data of the well site test and new data of the well site test according to the knowledge graph.
1. A processing method for data of a well site test based on a knowledge graph, comprising:
carrying out format identification on received historical data of a well site test, so as to generate format identification results;
establishing a mind map according to the format identification results;
generating a knowledge graph of the data of the well site test according to the mind map; and
processing the historical data of the well site test and new data of the well site test according to the knowledge graph;
the carrying out format identification on received historical data of the well site test to generate format identification results, comprising:
receiving and resolving binary streams and attached operation commands of files sent by a user through a network request; extracting information of file names, categories of operation objects and file formats from command parameters; scanning a target directory to determine whether folders corresponding to the extracted file formats exist or not; if a folder corresponding to the extracted file format is determined to be nonexistent, creating a new folder corresponding to the file format and writing a corresponding file under a newly created folder to generate a heterogeneous integrated database;
the establishing the mind map according to the format identification results, comprising:
determining keywords of the historical data of the well site test according to the format identification results;
establishing a data storage bank with a multi-level relationship according to the multiple keywords and a preset term dictionary of the well site test;
establishing the mind map according to the data storage bank;
the format identification results comprising: structured data, semi-structured data and unstructured data; the unstructured data comprising: a technical file, a picture/an audio/a video, an instrument and equipment ledger, actual drilling data;
the determining keywords of the historical data of the well site test according to the format identification results comprising:
carrying out grammar analysis on the structured data to determine keywords of the structured data;
calibrating labels of the semi-structured data and the unstructured data to determine keywords of the semi-structured data and the unstructured data;
the carrying out grammar analysis on the structured data to determine keywords of the structured data, specifically comprising: carrying out term extraction on the structured data according to the term dictionary of the well site test; selecting terms, a number of appearing frequencies of which is more than a preset number of times, from extraction results; generating feature vectors of the structured data according to the terms, the number of appearing frequencies of which is more than the preset number of times; and generating the keywords of the structured data according to the feature vectors;
the calibrating labels of the semi-structured data and the unstructured data to determine keywords of the semi-structured data and the unstructured data, comprising: calculating literal text similarities of the semi-structured data/the unstructured data and the term dictionary of the well site test; and selecting part of data from the semi-structured data and the unstructured data and calibrating the part of data according to the literal text similarities;
wherein a to-be-matched word B with the highest literal text similarity is selected as a label corresponding to data to be calibrated by the following formula, and the label is determined as the keyword to achieve extraction for the data of the semi-structured/unstructured data;
sim
=
60
×
(
xsword
ctrlword
+
xsword
keyword
)
/
2
+
40
×
dp
×
(
∑
c_xsword
(
i
)
ctrlword
(
i
)
+
∑
k_xsword
(
i
)
keyword
(
i
)
)
/
2
wherein sim represents the literal text similarity; xsword represents the number of same characters contained in a to-be-matched word A and a to-be-matched word B; ctrlword represents the total number of characters contained in the to-be-matched word A; keyword represents the total number of characters contained in the to-be-matched word B; dp represents a position coefficient representing a ratio of the total characters of the to-be-matched word A and the to-be-matched word B;
∑
c_xsword
(
i
)
∑
ctrlword
(
i
)
represents the sum of weight of the positions of the same characters contained in the to-be-matched word A and the to-be-matched word B in the to-be-matched word A;
∑
k_xsword
(
i
)
∑
keyword
(
i
)
represents the sum of weight of the positions of the same characters contained in the to-be-matched word A and the to-be-matched word B in the to-be-matched word B;
the generating a knowledge graph of the historical data of the well site test according to the mind map specifically comprising: carrying out granularity entity identification on the format identification results according to the mind map to generate identification results; establishing a knowledge level of the historical data of the well site test according to the identification results; extracting entity data of the historical data of the well site test according to the identification results; and generating the knowledge graph according to the knowledge level and the entity data.
2. An electronic equipment, comprising a memory, a processor and a computer program which is stored in the memory and can operate in the processor, wherein the processor is configured to achieve the steps of the processing method for data of a well site test based on a knowledge graph of claim 1 when executing the program.
3. A non-transitory computer readable storage medium, having stored thereon a computer program, wherein the computer program when executed by a processor achieves the steps of the processing method for data of a well site test based on a knowledge graph of claim 1 .
4. A processing device for data of a well site test based on a knowledge graph, comprising:
an identification result generation module, used for carrying out format identification on received historical data of the well site test to generate format identification results;
a mind map establishing module, used for establishing a mind map according to the format identification results;
a knowledge graph generation module, used for generating a knowledge graph of the historical data of the well site test according to the mind map; and
a data processing module, used for processing the historical data of the well site test and new data of the well site test according to the knowledge graph;
the carrying out format identification on received historical data of the well site test to generate format identification results, comprising:
receiving and resolving binary streams and attached operation commands of files sent by a user through a network request extracting information of file names, categories of operation objects and file formats from command parameters; scanning a target directory to determine whether folders corresponding to the extracted file formats exist or not if a folder corresponding to the extracted file format is determined to be nonexistent, creating a new folder corresponding to the file format and writing a corresponding file under a newly created folder to generate a heterogeneous integrated database;
the mind map establishing module comprising:
a keyword determination unit, used for determining keywords of the historical data of the well site test according to the format identification results;
a data storage bank establishing unit, used for establishing a data storage bank with a multi-level relationship according to the multiple keywords and a preset term dictionary of the well site test; and
a mind map establishing unit, used for establishing the mind map according to the data storage bank;
the format identification results comprising: structured data, semi-structured data and unstructured data; the unstructured data comprising: a technical file, a picture/an audio/a video, an instrument and equipment ledger, actual drilling data;
the keyword determination unit comprising:
a data grammar analysis unit, used for carrying out grammar analysis on the structured data to determine keywords of the structured data; and
a label calibrating unit, used for calibrating labels of the semi-structured data and the unstructured data to determine keywords of the semi-structured data and the unstructured data;
the data grammar analysis unit being specifically used for:
carrying out term extraction on the structured data according to the term dictionary of the well site test selecting terms, a number of appearing frequencies of which is more than a preset number of times, from extraction results; generating feature vectors of the structured data according to the terms, the number of appearing frequencies of which is more than the preset number of times; and
generating the keywords of the structured data according to the feature vectors;
the label calibrating unit being specifically used for:
calculating literal text similarities of the semi-structured data/the unstructured data and the term dictionary of the well site test and selecting part of data from the semi-structured data and the unstructured data and calibrating the part of data according to the literal text similarities;
wherein a to-be-matched word B with the highest literal text similarity is selected as a label corresponding to data to be calibrated by the following formula, and the label is determined as the keyword to achieve extraction for the data of the semi-structured/unstructured data;
sim
=
60
×
(
xsword
ctrlword
+
xsword
keyword
)
/
2
+
40
×
dp
×
(
∑
c_xsword
(
i
)
ctrlword
(
i
)
+
∑
k_xsword
(
i
)
keyword
(
i
)
)
/
2
wherein sim represents the literal text similarity; xsword represents the number of same characters contained in a to-be-matched word A and a to-be-matched word B; ctrlword represents the total number of characters contained in the to-be-matched word A; keyword represents the total number of characters contained in the to-be-matched word B; dp represents a position coefficient representing a ratio of the total characters of the to-be-matched word A and the to-be-matched word B;
∑
c_xsword
(
i
)
∑
ctrlword
(
i
)
represents the sum of weight of the positions of the same characters contained in the to-be-matched word A and the to-be-matched word B in the to-be-matched word A;
∑
k_xsword
(
i
)
∑
keyword
(
i
)
represents the sum of weight of the positions of the same characters contained in the to-be-matched word A and the to-be-matched word B in the to-be-matched word B;
the knowledge graph generation module being specifically used for:
carrying out granularity entity identification on the format identification results according to the mind map to generate identification results; establishing a knowledge level of the historical data of the well site test according to the identification results; extracting entity data of the historical data of the well site test according to the identification results; and generating the knowledge graph according to the knowledge level and the entity data.