Abstract—With the current trend of embedding location ser- vices within socialnetworks, an ever growing amount of users’ spatiotemporal tracks are being collected and used to generate user proﬁles. Issues of personal privacy and especially those stemming from tracking user location become more important to address. In this work, it is argued that support of locationprivacyawareness within socialnetworks is needed to maintain the users’ trust in their services. Current practices of pre-conﬁguring location disclosure settings have been shown to be limited, where users’ sense of locationprivacy dynamically change with context. In this paper, locationprivacyawareness is considered within a composite view of place, time and social data recorded in user proﬁles. The paper examines the possible threats to personal privacy from exposure of this data and the design of feedback tools to allow users to control their privacy. A user study is used to examine the impact of the feedback provided on users’ perception of privacy and the link between their privacy concerns and their attitude towards using the geo-social network. Findings conﬁrm the strong need for more transparent access to and control over user location proﬁles, and guide the proposal of recommendations to the design of more privacy-sensitive geo-socialnetworks.
Sharing location information on GeoSNs can pose several threats to user privacy. Providing location-based services requires the knowledge of the users’ location, and in some GeoSNs, this location information is very precise. Location is considered to be sensitive in comparison to other types of information. Location information typically includes the spatial (where) and temporal (when) factors that make it distinctive and dynamic in nature. Tracking a user’s loca- tion over time can reveal their identity [5, 6]. In addition, users’ historical location information can be linked to contextual and semantic information publicly available on GeoSNs and can be used for constructing comprehensive user profiles that include inferred personal information about users . Derived information in such profiles can include user activities, habits, mobility patterns and relationships with others [8, 9, 10, 11, 12]. Such enriched location-based profiles can be considered to be useful if used to personalise and enhance the quality and usability of the applications. However, it can potentially be used for undesirable purposes and pose new threats to users’ locationprivacy, which refers to a particular type of information privacy that sup- ports users’ right to be consent to all aspects of their location disclosure [13, 5]. Thus, location disclosure on GeoSNs can expand the implications to locationprivacy, compared to LBS. Users’ concerns about their locationprivacy are evident [14, 15, 16], yet the provided privacy solutions have shown to be ineffective. Privacy policies and privacy settings have been applied widely in online services as a means of providing privacy management to users by offering them some information about how their data are handled and basic access controls to their data. Nevertheless, privacy policies are relatively ineffective due to the vague and complex presentation of information . Privacy settings also enable limited protection that has shown to be difficult to use [18, 19]. In fact, users are unable to effectively use them due to their lack of awareness of the potential privacy consequences of their data exposure [20, 21, 22, 23]. In addition, LocationPrivacy-Preserving Mechanisms (LPPMs) have been developed to provide protection in LBS which mainly utilise anonymity and obfuscation techniques [13, 24]. However, these are not fully robust in protecting locationprivacy against attacks . They mainly work on reducing the spatial and temporal accuracy which restricts their applicability on GeoSNs and impacts the quality of service provided  (see Section 2.2.2). Hence, there is a need for effective locationprivacy solutions.
Also, static scenarios of the application use were adopted in the second experiment to gauge users’ perception and attitude to privacy. Whilst the users were initially primed and made aware of the prototypical nature of the test, the scenarios are admittedly limited in nature compared to realistic use of a working system. Thus the test environment may have influenced the choices users made in response to the questions asked. There are normally tradeoffs to be made when choosing a user-based experimental methodology, related for example, to the representativeness and size of sample used or the level of intrusiveness of the test. Hence, it is also important to employ other approaches in the future, for example, user observation or experience sampling, to compare and validate the results of this experiment. Future work will look further into the design aspects of the proposed feedback and control tools, in particular, the scope of information to be revealed and its timing with respect to task performance, and will seek in-depth evaluation of the usability aspects of the design.
Figure 4 shows the distribution of change in brokerage induced by the geo-social network with respect to each of the single-layer networks. For each node, this change is here measured as the difference between the effective size and ef- ficiency of the node’s neighbourhood in the composite geo- social network and the node’s effective size and efficiency in the single-layer network. When a node has a degree (and effective size) equal to zero in either layer, the node’s ef- ficiency is set equal to zero in that layer. Figure 4a shows changes in effective size only within the range (−50, 50). As suggested by the figure, there is an improvement of bro- kerage potential in the geo-social network over brokerage in the co-location network. When the social layer is also ac- counted for in the analysis of a node’s brokerage position, additional structural holes emerge in the node’s neighbour- hood, thus amplifying the node’s opportunities to intermedi- ate among disconnected others. However, while the majority of nodes can also improve their brokerage positions when the co-location layer is added to the social layer, nonethe- less there are some who suffer from a decrease in structural holes. Thus, the co-location network may contribute toward increasing the number of a node’s unique contacts, but at the same time may also add new links among some of the node’s contacts that would appear as unconnected in the so- cial layer. These mixed effects of the geo-social network on brokerage are even more pronounced when assessed in terms of variation in efficiency. As indicated by Figure 4b, while most nodes seem to secure a more efficient neighbourhood when one layer is combined with the other, there are some who suffer from a loss of efficiency, especially when the co- location layer is added to the social one.
long times with typical time intervals. This, nonetheless, significantly affects operation in addition to disconnects customers. The main element distinction among these solutions in addition to our own work is they depend upon dependable intermediaries, or maybe dependable machines, in addition to show rough realworld area towards machines with plain-text. Inside LocX, we all don't trust any intermediaries or maybe machines. For the beneficial aspect, these solutions are more normal in addition to, for this reason, can apply at many location-based solutions, even though LocX centers generally on the appearing geo-social applications. Your second group is area alteration, which usually utilizes developed area coordinates in order to preserve end user area privateness. Just one simple difficulty with finalizing nearest-neighbor queries using this type of approach is to effectively uncover each of the genuine neighbors. Shades assessment utilizing Hilbert Shape , however, can simply uncover rough neighbors. In order to find genuine neighbors, prior work either will keep the actual proximity connected with developed destinations in order to precise destinations in addition to incrementally operations nearest-neighbor queries or maybe demands dependable finally celebrations to do area alteration among customers in addition to LBSA machines. In contrast, LocX will not trust any finally get together and also the developed destinations aren't linked to precise destinations. Nonetheless, our bodies continues to be ready to look for the precise neighbors, and is particularly resistant in opposition to attacks depending on checking constant queries .
Open network structures and brokerage positions have long been seen as playing a crucial role in sustaining social capital and competitive advantage. The degree to which individuals intermediate between otherwise disconnected others can dif- fer across online and offline socialnetworks. For example, users may broker online between two others who then ex- change offline the information received through social media. Yet network studies of social capital have often neglected the interplay between online and offline interactions, and have concentrated primarily on a single layer. Here, we propose a geo-social multilayer approach to brokerage that casts light on the integrated online and offline foundations of social cap- ital. Drawing on a data set of 37,722 Foursquare users in London, we extend the notion of brokerage by examining users’ positions in an online social network and their offline mobility patterns through check-ins. We find that social and geographic brokerage positions are distinct and asymmetric across the social and co-locationnetworks. On the one hand, users may appear to be brokers online when in fact their abil- ity to intermediate would be mitigated if their offline posi- tions were also taken into account. On the other hand, users who appear to have little brokerage power offline may be ac- tive brokers within networks that combine both online and offline interactions. Our unified multilayer approach to bro- kerage enables us to uncover sources of social capital that would otherwise remain undetected if only the online or the offline layer were analysed in isolation.
We could cut down the size of predefined cells to increase the accuracy of proximity estimation. However, to do so, will reduce the privacy protection as well. If the size of cells is too small, then users’ locations might not be hidden to a sufficient level by the system. Finding well-defined geographical cells has been a critical problem. We propose that cells should be generated dynamically based on geographical factors and the distance between users. When two users are close to each other, the system should always be able to indicate that they are close and also be able to protect their locationprivacy. Cell towers are distributed in urban areas based on city planning and expected density of users. Moreover, each different cell tower has a different coverage range. Different combinations of cell towers can form different coverage shapes which increase the difficulty of executing a maximum movement boundary attack. A moving mobile user will be automatically connected to different cell towers that cover different regions. Therefore, using the coverage regions of surrounding cell towers as geographical cells can provide a dynamic grid for proximity testing. Two users who are close will always be in the same coverage. Thus, our proposal reduces the problem of false negatives and false positives.
Developing a model requires data on users’ privacy behaviour in opportunistic networks, but collecting such data is not straightforward. To collect high-quality data, it may be required to build, deploy, and measure user behaviour in a real, large-scale opportunistic network. But this may be time-consuming and impractical — and moreover, privacy behaviour in such an experimental network may not reflect actual behaviour, since users may be unfamiliar with these new technologies and so act in different ways . Thus, to develop our model, we instead measured privacy behaviour by performing a smaller-scale user study which investigated the location-sharing privacy preferences of 80 users of the popular online social network Facebook. 2
Nowadays, Geosocial networking application is using GPS location services to provide a social interface to the physical world. Android are quickly becoming the dominant computing platform for today’s user applications. Examples of popular social applications include social rendezvous , local friend recommendation for shopping and dining , , as well as collaborative network services and games , . The major problem is to design mechanisms that efficiently protect user privacy without sacrificing the accuracy of the system, or making strong assumptions about the security or trust worthiness of the application servers. More specifically, we target geo-social applications, and assume that servers (and any intermediaries) can be compromised and therefore, are untrusted. Mobile socialnetworks require stronger privacy properties than the open-to-all policies available today.
Encryption-based approaches to information hiding carry the risk that the OSN operator might enact a “no blobs of ciphertext” policy. Luo, Xie and Hen- gartner’s FaceCloak  avoids this particular risk by populating the user’s Facebook profile with plausible-looking data, while keeping real information on another server that only FaceCloak users are aware of. Lockr, by Tootoonchian et al.  uses social relationships defined by OSNs to control access to infor- mation stored outside the network: users share “social attestations”, tokens that represent a relationship, via a social network, and produce attestations to con- tent servers which check them against “Social ACLs”. Integration with services such as Flickr involve placeholders and external content storage services like FaceCloak. Guha, Tang and Francis’ NOYB  uses the clever approach of storing real data, in plaintext, on the profiles of other NOYB users; the location of a user’s actual data is given by a keyed permutation. NOYB does not address the key management problem.
The popularization of mobile communication devices and location technology has spurred the increasing demand for location-based services (LBSs). While enjoying the convenience provided by LBS, users may be confronted with the risk of privacy leakage. It is very crucial to devise a secure scheme to protect the locationprivacy of users. In this paper, we propose an anonymous entropy-based locationprivacy protection scheme in mobile socialnetworks (MSN), which includes two algorithms K-DDCA in a densely populated region and K-SDCA in a sparsely populated region to tackle the problem of locationprivacy leakage. The K-DDCA algorithm employs anonymous entropy method to select user groups and construct anonymous regions which can guarantee the area of the anonymous region formed be moderate and the diversity of the request content. The K-SDCA algorithm generates a set of similar dummy locations which can resist the attack of adversaries with background information. Particularly, we present the anonymous entropy method based on the location distance and request contents. The effectiveness of our scheme is validated through extensive simulations, which show that our scheme can achieve enhanced privacy preservation and better efficiency.
Throughout much of human history, technologies have been developed that make it easier for people to communicate . For example, the use of telegraph to transmit and receive messages over long distances dated back to 1792, . Emile Durkheim, a French sociologist known by many as the father of sociology, and Ferdinand Tonnes, a German sociologist are considered pioneers of socialnetworks during the late 1800s. Over the years, social media has evolved rapidly like LinkedIn which gained popularity in 2000. Youtube in 2005; then Facebook and Twitter by the year 2006 . Many researchers have worked tirelessly to enhance social media networks by building systems that will help users enjoy the usage of social media networks.
This paper presented a data-oriented approach to understanding users’ perception of threat to privacy on GeoSNs and proposes a model of privacy risks that is derived from studying collective users’ attitude to sharing data on GeoSNs. Aspects of the problem have been identified, namely, data, visibility and awareness. Data disclosed by sharing location information vary within a space defined by the spatial, temporal and socio-semantic dimensions. In addition, the sensitivity of the places visited as well as co-location with others were identified as contributing factors to privacy risks. An experiment has been designed to assess the effect of these variables on the privacy perception of GeoSN users. Perception of risk was noted to increase as the information content in the data disclosed increased; whether with the data dimension, or with place sensitivity and co-location with other people. Visibility of the information was shown to have the most significant impact on privacy perception, where users were more comfortable sharing their information with ‘Friends‘ (as defined by connections made on social web applications). Making the user aware of the nature of the information they are sharing was seen to have a significant impact on their sharing behaviour. This indicates that users’ awareness is limited when interacting on GeoSNs and thus questions their presumed consent of use of the applications.
Abstract: Research into the use of Location Based Services (LBS) that can pinpoint the exact location of users using wireless networks is the fastest growing area in Information Technology (IT) today. This is because of the need to transform the radio waves which act as a wireless networks data’s transmission medium into a private location. Contemporary research on LBS suggests that indoor location can be difficult as the geo positional satellites (GPS) cannot give an accurate positional computation due to insulation provided by physical barriers like the walls and furniture of a house. Previous research however suggests a way around this by making use of wireless fidelity (WiFi) cards signal strength but acknowledges limitations on the range which doesn’t exceed 50 meters. Other researchers have suggested that using LBS technology would allow hackers to track the user’s movement over time and so proposed that the user identity be kept secret by disposing the identifiers. Against this backdrop, some researchers have championed the call for a framework in LBS privacy in order to curtail the security risks that come with using wireless networks and suggested using a transactionbased wireless communication system in which transactions were unlinkable. This would in effect camouflage the movement of users as their location would not be able to be tracked. This paper aims to review contemporary issues on location based privacy in wireless technology and proposes a model for optimising LBS privacy and describes the initial stages of a research project aimed at filling the research void through the application of a hybrid research methodology
Abstract : Analyzing the conflict on privacy preserving mechanisms and functionality in geosocialnetworks, PROFILR-A framework is proposed for constructing location centric profiles (LCPs), aggregates built over the profiles of users that have visited discrete locations thereby preserving users from unwanted issues. PROFIL Rendows users with strong privacy guarantees and providers with correctness assurances. Steps are taken toward addressing this conflict. The approach is based on the concept of location centric profiles (LCPs). LCPs are statistics built from the profiles of users that have visited a certain location or a set of co-located users. A novel approach is proposed to define the location and user based safety metrics. Our key insight is to apply secure user-specific, distance-preserving coordinate transformations to all location data shared with the server. In addition to a venue centric approach, a decentralized solution is proposed for computing LCP snapshots over the profiles of co-located users is presented for private information retrieval that allows a user to retrieve information. In future, cryptographic techniques are further applied to enhance the security such that a technique from a database server without revealing what is actually being retrieved from the server. This allows all location queries to be evaluated correctly by the server, but our privacy mechanisms guarantee that servers are unable to see or infer the actual location data from the transformed data or from the data access.
plan is provably secure and thecryptographic capacities advanced for versatile stages however have computational overhead. Tor is programming that can beused for giving online anomity. Be that as it may, new research found that assaults can be made between the last transfer and the destination. Area to list mapping is another component which receives the area change from longitude convention furthermore presents part of area and information into two sections and putting away it in distinctive servers. The companions of a client share this present client's insider facts so they can apply the same change. This permits all area questions to be assessed effectively by the server, however our security systems ensure that servers are not able to see or surmise the real area information from the changed information or from the information access. This includes minimal computational and correspondence overhead to existing frameworks. Area to file mapping makes a major stride towards making area privacy practical for a large class of emerging geo-social applications.
We have seen that there is a wide variety of privacy issues that play a role in OSNs. Because the type of access differs greatly between users and Service Providers, the two main categories of threats require their own specific defense mechanisms. De- spite the fact that prevention is no simple matter, research is being conducted in many areas to alleviate some of the aforementioned threats. To protect user data from fellow users, awareness and proper tools for managing and enforcing access policies play a leading role [2, 12, 31]. This does not work towards solving issues that involve untrusted Service Providers. Obscuring and hiding sensitive data from the providers [1, 25, 48], or removing them from the picture entirely [8, 9, 47] are the general approaches here, as we will see. We now proceed to a topical literature overview of research on mitigating privacy issues and tailoring to the privacy needs of users.
One may argue that data collection is part of Facebook’s effort to increase user ex- perience. And some might find it acceptable that their data is collected and processed through face recognition, location tracking, and user behavior recording. But there are services and algorithms completely unrelated to socialnetworks that definitely don’t increase user experience within an OSN. For example, Facebook registered a patent for an algorithm calculating credit ratings using the social graph . That means if your friends are considered uncreditworthy, your own financial credibility is ranked down. Facebook may even calculate credibility through profile, image, and location analysis. If users do not want to be ranked based on their social profile, they cannot opt out or object without deleting their whole account.
can see that they are fully justified when thinking to the possible malicious uses of lo- cation information, such as robbing and stalking. For instance, the application “Girls Around Me”, combines social media and location information to find nearby women (who hadn’t necessarily agreed to be found), and, with one click the user can access the Facebook profiles of targeted girls . Particularly worrisome is the perspective of potential combination with the users’ most sensitive information, such as sexual ori- entation. Again, according to the Guardian , there have been cases of smartphone applications from which such information was collected without the user’s knowledge. Furthermore, location information can be easily used to obtain a variety of other information that an individual usually wishes to protect: by collecting and processing accurate location data on a regular basis, it is possible to infer an individual’s home or work location, sexual preferences, political views, religious inclinations, etc.
In this project, which we wish to discuss with a broad audience at GeoRich’15, we want to take the problem of spatial co-location mining into a new context, by considering spatio-temporal data, i.e., trajectory data of individuals. Thus, the problem now is to find groups of users which frequently co-locate in geo-space over time, creating the notion of geo-social co-location mining. There is already an abundance of public data sets that can be mined, in- cluding data sets from geo-socialnetworks  and from social net- works using geo-tags such as Twitter. Frequent co-location mining on such data may yield interesting patterns, such as “Members of LMU and HKU are frequently to be found at the same location, while members of some other university are often found in solitude or among themselves”. In such an application, each instance of a co-location corresponds to a (l, t, S) triple, where S denotes the set of individuals that have been at the same location l at the same time t. The problem of geo-social co-location mining introduces two major new challenges which have not been sufficiently covered in existing work on traditional co-location mining. Firstly, the tem- poral dimension leads to very large sets of co-location instances, since every location and time pair leads to a possibly non-empty co-location instance, secondly existing solutions do not consider the uncertainty which is inherent in spatial data: Spatial data may be imprecise (e.g., due to measurement errors), data can be obso- lete (e.g., when the most recent position update is already minutes old), data may originate from unreliable sources (such as crowd- sourcing), or it may be blurred to prevent privacy threats and to protect user anonymity . For example, the oval regions in Fig- ure 1 may correspond to individual persons, while the color of each person may represent the individual’s affiliations. Here, the loca- tion of each person is a conservative approximation based on the users GPS history. It is important to note that we are considering historic data. Thus, for a given point of time t, both past and future GPS positions of a user may be available. 1 Given these approxima-