os.chdir(Folder_Path)
file_list=os.listdir()
for i in range(1,len(file_list)):df=pd.read_csv(file_list[i],sep='\t')df.to_csv(SaveFile_Path+"/"+SaveFile_Name,encoding="utf_8_sig",index=False,header=None,mode='a+')sys.stdout.write("\r已合并：%.2f%%"%float((i/len(file_list))*100))sys.stdout.flush()

2、数据集处理

（1）查看空字段数量并排序

is_null=df_01.isnull().sum().sort_values(ascending=False)
is_null[is_null>row*0.85]#筛选出空值数量大于85%的数据

（2）删除缺失率大于85%字段

drop_columns=['Actor2Type3Code','Actor1Type3Code','Actor2Religion2Code',
'Actor1Religion2Code','Actor2EthnicCode','Actor1EthnicCode',
'Actor2Religion1Code','Actor2KnownGroupCode','Actor1Religion1Code',
'Actor1KnownGroupCode','Actor2Type2Code','Actor1Type2Code']
df_01.drop(drop_columns,axis=1,inplace=True)

得到结果如下：

3、数据导入MySQL与分析

（1）创建数据库

这里值得注意的是，很多字段在后续分析中没有用到，但还是导入进去了，为了和元数据保持一致性。

CREATE TABLE `export` (
`GLOBALEVENTID` int NOT NULL,
`SQLDATE` bigint,
`MonthYear` bigint,
`Year` bigint,
`FractionDate` bigint,
`Actor1Code` varchar(255),
`Actor1Name` varchar(255),
`Actor1CountryCode` varchar(255),
`Actor1Type1Code` varchar(255),
`Actor2Code` varchar(255),
`Actor2Name` varchar(255),
`Actor2CountryCode` varchar(255),
`Actor2Type1Code` varchar(255),
`IsRootEvent` varchar(255),
`EventCode` varchar(255),
`EventBaseCode` varchar(255),
`EventRootCode` varchar(255),
`QuadClass` int,
`GoldsteinScale` double,
`NumMentions` int,
`NumSources` int,
`NumArticles` int,
`AvgTone` double,
`Actor1Geo_Type` varchar(255),
`Actor1Geo_FullName` varchar(255),
`Actor1Geo_CountryCode` varchar(255),
`Actor1Geo_ADM1Code` varchar(255),
`Actor1Geo_ADM2Code` varchar(255),
`Actor1Geo_Lat` double,
`Actor1Geo_Long` double,
`Actor1Geo_FeatureID` varchar(255),
`Actor2Geo_Type` varchar(255),
`Actor2Geo_FullName` varchar(255),
`Actor2Geo_CountryCode` varchar(255),
`Actor2Geo_ADM1Code` varchar(255),
`Actor2Geo_ADM2Code` varchar(255),
`Actor2Geo_Lat` double,
`Actor2Geo_Long` double,
`Actor2Geo_FeatureID` varchar(255),
`ActionGeo_Type` varchar(255),
`ActionGeo_FullName` varchar(255),
`ActionGeo_CountryCode` varchar(255),
`ActionGeo_ADM1Code` varchar(255),
`ActionGeo_ADM2Code` varchar(255),
`ActionGeo_Lat` double,
`ActionGeo_Long` double,
`ActionGeo_FeatureID` varchar(255),
`DATEADDED` bigint,
`SOURCEURL` text,
PRIMARY KEY (`GLOBALEVENTID`)
);

（2）导入数据

LOAD DATA INFILE 'E:/term/code/mergedata/export/export_202201.csv' INTO TABLE
`export`
FIELDS TERMINATED BY ','
LINES TERMINATED BY '\r\n'
IGNORE 1 ROWS;

（3）年份处理和简单查询

DELETE
FROM rus_and_ukr
WHERE `MonthYear`<202201
SELECT COUNT(GLOBALEVENTID) FROM export -- 21504131

接下来将针对俄乌冲突进行筛选查询和可视化~