Location privacy and public metadata in social media platforms: attitudes, behaviors and opinions
Summary (4 min read)
1 Introduction
- Many advanced information systems may use geolocation technologies to identify user location in order to provide customized services by delivering localized news, recommending friends, serving targeted ads, improving large scale systems (e.g., cloud computing, content-based delivery networks).
- The knowledge of who is in the area may give access to real-time hazards, disaster information, photos and video that would likely speed-up the rescue operations.
- User’s geographical location can be generated either by users (through voluntary check-in in applications like Foursquare and Facebook Places) or by applications (through technologies like IP address geolocation, cellphone network triangulation, RFID and GPS).
- Moreover, in addition to geolocation information, many applications attach to users’ contents a lot of other information like OS language, device type and capture time.
- Results obtained from the first phase show that people are not concerned about privacy, but, in the second phase, these people are the most worried about the location-aware scenario; conversely, people who were initially concerned, are less worried about the location-aware scenario and find the scenario interesting.
3 Users’ attitudes toward privacy in location-based applications
- Unfortunately, as shown in the previous section, some studies say users are concerned about privacy, while other studies say they are not.
- In particular, the authors are interested in attitudes and behaviors of people who daily use technological devices and mobile applications, who consider mobile devices and applications as commodities and not as technological pieces of hardware to be scared of.
- To get in touch with these people, the authors asked for voluntary participants through different technological platforms (social networks, emails and forums) and they did not specify any age limit.
- The authors have been contacted by 122 people (2 to 6 times the number of participants to other studies in the field) and Figure 1 reports the charac- teristics of the 122 participants: 54% are women and 46% are men.
3.1 Technological equipment and habits
- One of the goals of the questionnaire was to understand the technological equipment (smartphone penetration, availability of GPS technology) and the users’ habits in the mobile scenario (data subscription plan, download of mobile applications and usage of geolocation services).
- Therefore, access to the Internet through smartphone devices is very common among respondents.
- Looking at the people who keep the GPS active, the authors can observe a different behavior between men and women: 61% of men keep the GPS on, but the percentage decreases to 37% when analyzing women.
- To better understand the behavior of the respondents with respect to the use of location-aware applications, Figure 5 shows the behavior of people who keep the GPS on and the behavior of those who have no idea about GPS availability over their devices.
- Results also show that people have a poor grasp of technologies/applications they use.
3.2 Twitter and Instagram Presence
- The second main goal of the questionnaire was to understand how users share their contents in social platforms.
- Figure 9 reports results obtained while asking participants the profile they use on Twitter.
- Similar results can be observed in Figure 10, where the authors report results obtained while asking participants the profile they have on the Instagram platform: 59% of men and 68% of women have no idea about the nature of their profile.
- It can be observed that even among users with a private profile, the use of the geolocation feature is considerable (35% in Twitter and 43% in Instagram).
- Again, this means that most of the users generated contents (textual or multimedia) are publicly accessible and contain geographical information.
3.3 Summary of Results
- The authors aimed at understanding if people are aware of the characteristics of the devices and applications they use.
- In summary, the first part of the investigation depicts the following scenario.
- The majority of the respondents claim to use location-aware mobile applications.
- This may represent a privacy risk for users; – Use of the geolocation feature.
- To really understand users’ opinions and preferences of this novel scenario, the authors think it is necessary to show them what third parties applications can do by browsing data in public social media platforms.
4 Location-aware scenario: an example
- The real-world study presented in the previous section showed that people are not really concerned about privacy and largely use geolocation technologies.
- The authors show how to develop a simple application able extract personal and sensitive users information from contents publicly available in Twitter and Instagram and able to use these information to locate, in real-time, users on a map.
- Note that the authors focus their attention on the Twitter and Instagram platforms for two main reasons: i) it is possible to browse users’ generated contents without signing up for the service, and ii) both platforms present a high volume of public data (no need to be friend to see someone’s tweets or photos).
- The application does not infer or compute any user’s location, but it relies on the location available in the metadata associated to user generated contents.
- In the following, the authors present details of the application development and of the application output.
4.1 Application Development
- The application is developed with Python 2.72 and, as depicted in Figure 18, is logically organized into four main blocks: i) Geocoding, ii) Tweets and photos Retrieval, iii) Users Filtering, and iv) KML production.
- The application uses the geopy library3 to convert a textual address (street name and city) into its GPS coordinates.
- The application directly interacts with Twitter and Instagram platforms through the platforms’.
4.2 Application Input/Output
- The application is developed to show that anyone can exploit social platforms APIs to develop a program able to browse for geotagged contents (i.e., for contents generated in a specific neighborhood) and able to retrieve personal and sensitive information from each geotagged contents.
- It has a textual interface and requires: an address, the radius where to search for users and the filtering options (see User Filtering Module).
- Figure 19 shows how Google Earth displays KML file produced by the developed application: all the users located in the specific area (Piazza Maggiore, Bologna, Italy) are described through yellow pins and when a specific user is selected, the retrieved information about the user are displayed.
- Needless to say, the authors also know she was in Galleria Cavour (a shopping center near Piazza Maggiore square in Bologna) when they checked the area (on July 11, 2013).
5 Location-aware scenario: opinions and preferences
- In the previous section, the authors showed that a simple application may access to personal and sensitive data without asking for permission and without logging-in into the social platforms.
- With respect to the Twitter scenario, reactions of Instagram users are more negative and settled.
- Men who did not change their privacy settings require third parties to be authorized (39%).
5.1 Summary of Results
- The second part of the investigation aimed at understanding users’ opinions and preferences about a location-aware scenario after seeing what a simple application can do by browsing data publicly available in social platforms.
- In summary, the second phase of the investigation depicts the following scenario.
- When talking about photo the “angry” reaction is predominant.
- Men find the service “Interesting”, whereas women find it “Intrusive”.
- If on the one side users are worried about the usage of public data to locate and contact them, on the other side they did not have preconceived ideas about a location-aware scenario.
6 Guidelines for developing an effective location-aware scenario
- Results showed that technology is not a burden for the development of an effective location-aware scenario, but also highlighted that users are not completely aware of what happens in the mobile scenario when producing or sharing contents.
- Similarly to the message that pops-up the first time the device tries to go on-line (“Continuing Internet access will lead to traffic.
- A message like “The photo you are sharing allows third parties to locate you” allowing users to select between two possible options “Remove geolocation info” or “Continue” would be very informative for the users.
- From the user point of view, a successful location-aware scenario will facilitate many aspects of their lives, but it will also represent a risk for their privacy.
7 Conclusions
- The authors investigated attitudes, behaviors and opinions of users with respect to the location-aware scenario.
- The investigation was done in two phases: the first phase aimed at understanding what people know or ignore of a location-aware scenario and the second phase investigated users’ opinions after showing them the amount of personal and sensitive information that a simple application can access to by browsing contents publicly available in social media platforms.
- Results also showed that men are more willing than women to enter the location-aware scenario, but both require to give authorization and to receive benefits when third parties access to their contents.
- Other interesting findings were that users do not want to be bothered with marketing or advertising services, that women are very alarmed if third parties would access to their photos and that photos are considered as a personal resource that need to be protected from third parties access.
- The analysis of the obtained results allowed us to outline possible guidelines that the authors think might be helpful for both users and developers/enterprises to build an effective location-aware scenario.
Did you find this useful? Give us your feedback
Citations
85 citations
Cites background from "Location privacy and public metadat..."
..., personal information in exchange of an email account), and they are increasingly interested in their privacy [36,37]; the lack of clarity about who has access to data may limit the growth of the IoT scenario....
[...]
65 citations
Cites background from "Location privacy and public metadat..."
...Moreover, the remaining vehicles will introduce the period of silence, exposing less information for the adversary, when vehicles change their pseudonyms [14, 21, 38, 42, 69, 71]....
[...]
44 citations
43 citations
Cites background from "Location privacy and public metadat..."
...A group could be formed from several consumers by applying any of the existing grouping schemes, such as spatial and temporal grouping schemes [13, 14]....
[...]
35 citations
Cites background from "Location privacy and public metadat..."
...There is a large variety of users posting messages (from ordinary individuals to important people, from public to private organizations, etc.) and tweets contain a wide variety of information, ranging from relevant to useless [7], [8], [9]....
[...]
References
2,922 citations
1,903 citations
1,213 citations
785 citations
"Location privacy and public metadat..." refers background in this paper
...possibilities are so broad that location-aware social applications are likely to be important for the next generation mobile computing [1]....
[...]
605 citations