The processed data are also published, including the observed past locations and lifetime durations. The historical data used in this work are made available as a single collection for result replication and further work in, with the list of files and their format described in Appendix A. To the best of my knowledge, it is the only universal solution for retrospective location. This makes their past versions available in data archives such as OS repositories and software development snapshots on GitHub and GitLab. This database is commonly shipped with various software and operating systems. The analyzed locations were observed via the past MaxMind GeoLite2 City databases. The results dealing with IP address location lifetime ( Section 6) are also valid for mobile devices. Therefore, these results are mostly descriptive of Europe and North America. In addition, the historical ground truth is not evenly distributed across the world. The results dealing with retrospective IP address locations ( Section 5) are limited to fixed devices. The presented results have a defined scope. The fitted lifetime model allows for a given address to estimate the time when its location was changed in the past. Specifically, interval censoring is used to cover the past dates where the historical geolocation databases are missing. Due to missing data, survival analysis is applied to process the data. For this purpose, I evaluate the database-stored location history of one year of about 421 k IPv4 and 47 k IPv6 addresses. I also work with the location lifetime in a database. The results show how far into the past the system may locate an address without a noticeable accuracy loss. The NA (not available not returned) locations are also considered in the analysis as their number significantly affects the accuracy, especially for IPv6 addresses. This ground truth was linked to historical geolocation databases to observe the past address locations. 51 k IPv4 and 17 k IPv6 addresses in the ground truth. I work with the historical ground truth that includes past IPv4 and IPv6 addresses and their locations over five years. The stored locations in a database are shared by a range of addresses to maximize the Internet space coverage. Such databases are populated by various techniques, which include location self-reporting, network measurements, mining web content, host and domain-name analyses, and custom submissions. Historical IP geolocation databases are used to obtain past locations. In a secured system, such as an e-shop with stored credit card details for one-click payments, a confident suspicion of ID theft can prevent the payment to minimize fraud losses and chargebacks. The history of address locations is used to estimate the confidence of the user travel between places of subsequent logins. The second use case discusses the application in identity theft prevention. The past viewers’ locations are used to investigate the reasons for the page loading errors. The first deals with the geotargeted online content in which some pages with dynamically generated content based on the viewers’ locations do not work for unknown reasons. The usage of the results is demonstrated by two longitudinal use cases. It also introduces an approach to handle missing historical data. This work presents results of retrospective location. However, this is in reality not true as only pieces of historical data are available, which makes the retrospective location a challenge. In theory, there is an unlimited history of all IP address locations available. These usages include evaluation of longitudinal studies, observation of long-term location patterns, replication of past system states, study of long-term evolution of the global Internet, and investigation of crime incidents. Any of these locations may be needed retrospectively when the reason to locate the device was not known before, or when the locations were obtained but not archived. It delivers the geographical location of any Internet device, independent of its use, installation, software, and hardware. IP geolocation is a fundamental part of many Internet services and applications.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |