Twitter and other social media are basically through the metadata (Metadata) to extract and save some personal information, in fact, from a 140-character message can get a lot of useful metadata. All of these metadata can be accessed through the open Twitter API.Here are some metadata examples, and anyone (not just the government) can use this information to “fingerprint” or track someone:
Average tweet activity, by hour and by day of the week
Timezone and language set for the Twitter interface
Sources used (mobile application, web browser, …)
Most used hashtags, most retweeted users and most mentioned users
Friends analysis based on most frequent timezones/languages
It is estimated that everyone is aware of the impact of leakage on the privacy of individuals, but also very few people realize that some regular frequent tweets can also send personal habits and other information. A single tweet may contain only some interesting metadata information, but thousands of tweets may reveal some unique personal lifestyle, which is interesting.
Meta-data form the structure and semantics of the meta-model, and the instantiated set of meta-data is called meta-model. After collecting a large amount of tweets, we can distinguish between “business accounts” which are “personal accounts”, but also to identify which users have an interactive relationship. To this end, for a particular account, x0rz github user wrote a Python script, which detects the frequency, time zone and language, location, tweets tag, forwarding account, friend interaction and other information, with access to the latest tweets, Take metadata, and identify every hour of Twitter usage every day. The following are two examples of the analysis.
An analysis of the Snowden Twitter account @Snowden
Snowdon has released 1682 tweets since September 2015, and has been able to find information about his daily and weekly tweets sending time, frequency, and sleep patterns.
Analysis of the Trump Twitter account @realdonaldtrump
Trump’s Twitter account is managed by many people? So many tweets send, it is worth recalling!
Twitter account analysis tool source code download: http://ift.tt/2kimz54
usage: tweets_analyzer.py -n [options]
Simple Twitter Profile Analyzer
-h, –help show this help message and exit
-l N, –limit N limit the number of tweets to retreive (default=1000)
-n screen_name, –name screen_name
-f FILTER, –filter FILTER
filter by source (ex. -f android will get android
–no-timezone removes the timezone auto-adjustment (default is UTC)
manually apply a timezone offset (in seconds)
–friends will perform quick friends analysis based on lang and
timezone (rate limit = 15 requests)