The document discusses several topics: how mouse movement and errors on unexpected questions can distinguish liars from truth-tellers in machine learning classifiers; how analyzing smartphone location data between 1am-4am and 1pm-5pm can identify people's homes and likely Thanksgiving locations; and how increased political advertising can shorten Thanksgiving celebrations, especially for those exposed to more ads or with different political views than their hosts. It also briefly mentions that WiFi signal strength can estimate device positions with centimeter accuracy.
28. Parameters that encode mouse movement were analyzed using
machine learning classifiers and the results indicate that the
mouse trajectories and errors on unexpected questions
efficiently distinguish liars from truth-tellers. Furthermore, we
showed that liars may be identified also when they are
responding truthfully. Unexpected questions combined with
the analysis of mouse movement may efficiently spot
participants with faked identities without the need for any
prior information on the examinee.
29. By downloading where these
smartphones were located between the
hours of 1 a.m. to 4 a.m., Chen and his
research partner — Ryne Rohla of
Washington State University — could
infer where people in this database
lived. To be safe, they tracked these
locations over the three weeks leading
up to Thanksgiving and made sure the
readings were consistent.
Doing so identified 6.3 million probable
homes. Repeating this computational
trick for the hours of 1 p.m. to 5 p.m. on
Thanksgiving revealed the places where
people most likely spent their festivities.
The team linked these home locations to precinct voting records, then made a big assumption. With these political leanings in hand, the team looked at how people from Democratic-leaning
homes behaved when they visited a Republican household for Thanksgiving and vice versa. Chen and Rohla conducted this analysis for Thanksgiving in 2015 and 2016 to see how behaviors
differed before and after the election.
Overall, members of both parties shortened their holiday trips by 30 to 50 minutes, though the team spotted subtle differences. Democrats were more likely to forgo Thanksgiving travel
altogether in 2016. But GOP supporters were more likely to leave dinner early, spending about four percent less time at Thanksgiving than those across the aisle. Thanksgiving dinners are
shortened by around 2.6 min on average for every 1,000 political advertisements aired in the traveler’s home media market. This selective exposure — the tendency to avoid people who
have different views — increased dramatically if folks hailed from a voting precinct exposed to tons of political ads. People from areas with a lot of political advertising cut close to 1.5 hours
from their Thanksgiving celebrations. The study found that “Thanksgiving dinners are shortened by around 2.6 min on average for every 1,000 political advertisements aired in the traveler’s
home media market.”
30.
31.
32.
33.
34. Meraki Access Points also detect the
signal strength of data frames and
probe requests, which can be used
to estimate the physical position of
the WiFi devices.
In this paper, we propose an indoor positioning system
which achieves centimeter accuracy and maintains the
performance under non-line-of-sight scenarios using a
single pair of off-the-shelf WiFi devices. By harvesting
the inherent diversity in WiFi systems, the proposed
indoor positioning system formulates a large effective
bandwidth to provide a fine-grained location-specific
fingerprint with a much higher resolution compared
with the fingerprints in most WiFi-based indoor
positioning systems.
https://code.org/curriculum/course2/18/followthedigitaltrail.pdf
We are surrounded by sensors. This sensors are collecting data about us and our environment. With wide avaliablity of sensory data, some creative thinking and use of machine learning and artificial intelligence alghoritms we can start discovering stories which are far beyond original intention of sensors which collected the data. This data can quite often be used to draw completely unexpected and unintended conclusions about us, our behaviour and habbits. They invade our privacy in ways previously unthinkable and to most people completely uncomprihensible.
Jet lag or inflamation (example)
Cocaine dileri koji dijele fitbite da paze da se ekipa ne predozira
Surveillance footage near the Aiello’s house stood in conflict with Tony Aiello saying his stepdaughter drove by with an unidentified passenger. But a security camera facing Navarra’s driveway from a neighbor’s home did capture Tony’s car parked at her house that afternoon. The motion-activated recordings caught glimpses of Tony’s car at Navarra’s house on Saturday, September 8 at 3:12, 3:14, 3:19, 3:21, and 3:33pm. A clip at 3:35 showed the car gone.
Meanwhile, Fitbit’s Bonham got back to the investigators with the health data. He reported that Navarra's device had been connected via Bluetooth to a nearby paired device when police discovered her, and the Fitbit had been reporting data every 15 minutes. The investigators noted that Navarra’s desktop computer was just five to ten feet away from where they found her body in the dining room. Her last recorded movement was on Thursday, September 13, approximately when the coroner removed her body from her home.
Before that, her last movement was on Saturday, September 8, the day Tony dropped off the pizza. It was also the last day the device recorded her heart rate. The Fitbit recorded a “significant” heart rate spike at 3:20pm, and it then rapidly declined. By 3:28—while Tony’s car was still parked in her driveway—her heart had stopped beating, according to the device.
Here below I have tried to step-by-step illustrate how Google shares location data when "legally" required:The authorities reached out to Google with a geofence warrant looking for smartphones Google had recorded around the crime scene.
After receiving the warrant, Google gathers location information from its Sensorvault database and sends it to investigators, with each device identified by an anonymous ID code and not the actual identity of the devices.
Investigators then review the data, look for patterns of the devices near the crime scene, and request further location data on devices from Google that appear relevant to see the particular device movement beyond the original area defined in the warrant.
When investigators narrow results to a few devices, which they think may belong to suspects or witnesses, Google reveals the real name, email address and other data associated with the devices.
Investigators use DNA, genealogy database to ID suspect in 1987 double homicide
Digitalna DNA forenzika, retroaktivno funkcionira
The Future of Crime-Fighting Is Family Tree Forensics
https://www.wired.com/story/the-future-of-crime-fighting-is-family-tree-forensics/
Want to figure out if someone is a psychopath? Ask them what their favourite song is. A New York University study last year found that people who loved Eminem’s Lose Yourself and Justin Bieber’s What Do You Mean? were more likely to score highly on the psychopathy scale than people who were into Dire Straits.
As ad targeting gets ever more sophisticated, marketers will have the ability to target our emotions in potentially exploitative ways. According to one study, titled Misery Is Not Miserly, you are more likely to spend more on a product if you’re feeling sad. You can imagine some companies might take advantage of that.
Kosher or halal, vegetarian or not, how often do you restock your fridge, when do you eat
Nije svaki IOT nužno dobar ili poželjan, tek toliko da svima bude jasnosamo zato što možemo nešto spojiti na Internet to ne znači da je to i dobra ideja
Surge washing, pitanje trenutka
Kud je roomba otputovala da je na rubu litice?
It suggests that insurance companies might use the smart meter data to determine health care premiums, such as if there is high usage at night which would indicate sleep behavior problems. Besides looking to bust pot farmers, law enforcement might use the data as "real-time surveillance to determine if residents are present and current activities inside the home." The press might wish to see the smart meter data of celebrities. Criminals may want to see the data to determine the best time for a burglary and what high dollar appliances you might have to steal. Marketers might want the data for profiling and targeting advertisements. Creditors might want the data to determine if behavior indicates creditworthiness.
To train their platform to identify different steps from local individuals, the researchers used eight barefoot participants and collected the time and frequency of their footfalls, as well as their cadence and length. Over a one-month period, they used their system to collect 46,000 footfall events from the participants and found that it took only eight minutes (or roughly 875 footsteps) for the system to identify specific individuals at an accuracy of 92.29%.
We need 130 servers and 1MW of power to turn your heater on or offs
Temperature, we know if you are man or a woman and who is in control of the thermostat.
If co2 rises above some level decision making becomes impairedbased on known facts how big is the space, we can calculate how many people are occupying the space
50 lica po jednom frameu (dakle 3000 lica u sekundi), za kameru
Findface je ruski softver, 40 pixela udaljenosti između zjenica za prepoznavanje osobe
China Airport face recognition systems to help you check your flight status and find the way to your gate. Note I did not input anything, it accurately identified my full flight information from my face!
They analyzed 22 manufacturers' products -- 1,294 in all -- spanning 4,956 firmware versions spread across 3,333,411 binaries.
Vendors are failing to implement basic hardening features, including decades-old best practices.
Many of these devices are low cost/high volume. They are given the minimal amount of development time to get a new product out the door. This means that security likely finds itself low on the list of priorities.
In the case of home routers and IoT devices, these devices sit in a location of privilege within the users’ home networks. Regardless of whether or not the device’s owner is an intended target, as “set it and forget it” appliances, these devices are an ideal hiding spot for botnets and other attacker-infrastructure. In short, it is a Good Idea for these devices to be reasonably secure.
Unfortunately, if the trend of minimal hardening does not change, these devices will continue to be a soft target for these types of activity.
The prevalence of duplicate binaries indicates that it might be possible to fix some issues at a single point. In the case of Buildroot, this can even be done by the community through pull requests.
Surprisingly, we found that 3,704 binaries were duplicated between at least two vendors. Some binaries were found in 9 different vendors.
Duplication between products was also very common; some binaries were duplicated between 109 different products.
https://cyber-itl.org/2019/08/26/iot-data-writeup.html
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0177851
https://en.wikipedia.org/wiki/Fitts's_law
prototype detection system for computer-mediated deception when face-to-face interaction is not available." They say that in a series of experiments, they were able to train a machine learning model to separate liars and truth-tellers by watching a one-on-one conversation between two people typing online, while using only the content and speed of their typing—and none of the other physical clues that polygraph machines claim can sort lies from truth.
The team linked these home locations to precinct voting records, then made a big assumption. With these political leanings in hand, the team looked at how people from Democratic-leaning homes behaved when they visited a Republican household for Thanksgiving and vice versa. Chen and Rohla conducted this analysis for Thanksgiving in 2015 and 2016 to see how behaviors differed before and after the election.
Overall, members of both parties shortened their holiday trips by 30 to 50 minutes, though the team spotted subtle differences. Democrats were more likely to forgo Thanksgiving travel altogether in 2016. But GOP supporters were more likely to leave dinner early, spending about four percent less time at Thanksgiving than those across the aisle. Thanksgiving dinners are shortened by around 2.6 min on average for every 1,000 political advertisements aired in the traveler’s home media market. This selective exposure — the tendency to avoid people who have different views — increased dramatically if folks hailed from a voting precinct exposed to tons of political ads. People from areas with a lot of political advertising cut close to 1.5 hours from their Thanksgiving celebrations. The study found that “Thanksgiving dinners are shortened by around 2.6 min on average for every 1,000 political advertisements aired in the traveler’s home media market.”
Facebook knows when you are about to enter relationship. Approximately 100 days before changing their status to “In a relationship,“ users are more likely to use the words ”love,“ “sweet,” and ”happy" in messages and publish something on the walls of their crushes — an average of 2 posts per 12 days.
Facebook / društvene mreže znaju kada se svađamo
About a week before the relationship is over, the user begins to communicate more actively with friends and family: exchanging messages and leaving comments under publications. On the day of the breakup, the number of interactions with other users increases by 225%: this is support from friends and relatives rolling in.
Naravno tu je i aplikacija za to
Cisco meraki
https://ieeexplore.ieee.org/document/7820842/
In Retail, People Counting & Anonymous Tracking solutions are used to compare store performance, mall analytics, and in-store optimization to increase conversions & profits.
When a person is behind a wall, then part of this WiFi signal is again reflected back through the wall and can subsequently be detected by a receiver. This technique is imminently suitable for the detection of moving people, because these generate ever changing reflection patterns, which are easily distinguishable from stationary objects.
The algorithm was trained using more than 150,000 satellite images of six cities, as well as 96 categories of points of interest like grocery stores and pet shops. It was then correlated with obesity rates reported from each city. The researchers included the points of interest because they could have an effect on the activity of a neighborhood. An area with pet shops could have more people taking dogs for a walk, for instance.