Niharika Sachdeva

A three-way investigation of a game-CAPTCHA: automated attacks, relay attacks and usability

Existing captcha solutions on the Internet are a major source of user frustration. Game captchas are an interesting and, to date, little-studied approach claiming to make captcha solving a fun activity for the users. One broad form of such captchas -- called Dynamic Cognitive Game (DCG) captchas -- challenge the user to perform a game-like cognitive task interacting with a series of dynamic images. We pursue a comprehensive analysis of a representative category of DCG captchas. We formalize, design and implement such captchas, and dissect them across: (1) fully automated attacks, (2) human-solver relay attacks, and (3) usability. Our results suggest that the studied DCG captchas exhibit high usability and, unlike other known captchas, offer some resistance to relay attacks, but they are also vulnerable to our novel dictionary-based automated attack.

Online Social Networks and Police in India—Understanding the Perceptions, Behavior, Challenges

Safety is a concern for most urban communities; police departments bear the majority of responsibility to maintain law and order and prevent crime. Police agencies across the globe are increasingly using Online Social Network (OSN) (such as Facebook and Twitter) to acquire intelligence and connect with citizens. Developing nations like India are however, still exploring OSN for policing. We interviewed 20 IPS officers and 21 citizens to understand perceptions, and explored challenges experienced while using OSN for policing. Interview analysis, highlights how citizens and police think about information shared on OSN, handling offensive comments, and acknowledgment overload, as they pursue social and safety goals. We found that success of OSN for policing demands effective communication between the stakeholders (citizens and police). Our study shows that OSN offers community-policing opportunities, enabling police to identify crime with the help of citizens. It can reduce the communication gap and improve coordination between police and citizens. We also discuss design opportunities for tools to support social interactions between stakeholders.

Call for Service: Characterizing and Modeling Police Response to Serviceable Requests on Facebook

Social media platforms have obtained substantial interest of police to connect with residents. This has encouraged residents to report day-to-day law and order concerns such as traffic congestion, missing people, and harassment by cops on these platforms. In this paper, we study day-to-day concerns shared by residents on social media and police response to such concerns. Based on the input of police experts, we define concerns that require police response and attention, as a serviceable request. We provide insights on six textual attributes that can identify serviceable posts. We find such posts are marked by high negative emotions, more factual, and objective content such as location and time of incidences. We show that police response time varies depending upon the kind of serviceable requests. Our work explores a series of statistical models to predict serviceable posts and its different types. We conclude the paper, discussing the implication of our findings on police practices and design needs for possible technological interventions. These technological interventions will help increase the interactions between police and residents and thereby increasing the well-being and safety of society.

On the security and usability of dynamic cognitive game CAPTCHAs

Existing CAPTCHA solutions are a major source of user frustration on the Internet today, frequently forcing companies to lose customers and business. Game CAPTCHAs are a promising approach which may make CAPTCHA solving a fun activity for the user. One category of such CAPTCHAs – called Dynamic Cognitive Game (DCG) CAPTCHA – challenges the user to perform a game-like cognitive (or recognition) task interacting with a series of dynamic images. Specifically, it takes the form of many objects floating around within the images, and the user’s task is to match the objects corresponding to specific target(s), and drag/drop them to the target region(s). In this paper, we pursue a comprehensive analysis of DCG CAPTCHAs. We design and implement such CAPTCHAs, and dissect them across four broad but overlapping dimensions: (1) usability, (2) fully automated attacks, (3) human-solving relay attacks, and (4) hybrid attacks that combine the strengths of automated and relay attacks. Our study shows that DCG CAPTCHAs are highly usable, even on mobile devices and offer some resilience to relay attacks, but they are vulnerable to our proposed automated and hybrid attacks.

Social Media - New Face of Collaborative Policing?

Online social media (OSM) has become a preferred choice of police to communicate and collaborate with citizens for improved safety. Various studies investigate perceptions and opinion of high ranked police officers on use of OSM in policing, however, understanding and perceptions of field level police personnel is largely unexplored. We collected survey responses of 445 police personnel and 204 citizens’ survey in India to understand perceptions on OSN use for policing. Further, we analyzed posts from Facebook pages of Indian police organizations to study the behavior of police and citizens as they pursue social and safety goals on OSN. We find that success of OSN for policing demands effective communication between the stakeholders (citizens and police). Our results show preliminary evidences that OSN use for policing can help (1) increase participation in problem solving process, (2) increase community engagement by providing unique channel for both Feedback and Anonymity. However, such a system will need appropriate acknowledgment and trustworthiness channels to be successful. We also identify challenges in adopting OSN and outline design opportunities for HCI researchers and practitioners to design tools supporting social interactions for policing.

On the viability of CAPTCHAs for use in telephony systems: a usability field study

Usability of security solution has always been a keen area of interest for researchers. CAPTCHA is one such security solution which presents various usability challenges for users. However, it has successfully reduced the abuse of the Internet resources, such as spam. Similar to the Internet, audio-based CAPTCHAs have been proposed as a solution to curb voice spam over telephony. Voice spam is often encountered on telephony in various forms, such as, an automated telemarketing call asking to call a number to win million of dollars. A large percentage of voice spam is generated through automated system which introduces the classical challenge of distinguishing machines from humans on the telephony. We present a large scale evaluation of audio CAPTCHA from the human perspective over telephony through a field study with 90 participants. We study two primary research questions: how much inconvenience does audio CAPTCHA causes to users on telephony, and how different features of the CAPTCHA, e.g., duration and size influence usability of audio CAPTCHA on telephony. We found that captcha could be a viable solution for telephony with improved features, such as better voice and accent. We found that users were relatively close to the expected correct answers, which does suggest the possibility of deploying audio captcha on telephony platforms in the future. However, we did not find strong influence of captcha size and duration on solving accuracy.

Stop the KillFies! Using Deep Learning Models to Identify Dangerous Selfies

Selfies have become a prominent medium for self-portrayal on social media. Unfortunately, certain social media users go to extreme lengths to click selfies, which puts their lives at risk. Two hundred and sixteen individuals have died since March 2014 until January 2018 while trying to click selfies. It is imperative to be able to identify dangerous selfies posted on social media platforms to be able to build an intervention for users going to extreme lengths for clicking such selfies. In this work, we propose a convolutional neural network based classifier to identify dangerous selfies posted on social media using only the image (no metadata). We show that our proposed approach gives an accuracy of

PicHunt: Social Media Image Retrieval for Improved Law Enforcement

98%

On the Viability of CAPTCHAs for use in Telephony Systems: A Usability Field Study

and performs better than previous methods.

Home is safer than the cloud!: privacy concerns for consumer cloud storage

First responders are increasingly using social media to identify and reduce crime for well-being and safety of the society. Images shared on social media hurting religious, political, communal and other sentiments of people, often instigate violence and create law & order situations in society. This results in the need for first responders to inspect the spread of such images and users propagating them on social media. In this paper, we present a comparison between different hand-crafted features and a Convolutional Neural Network (CNN) model to retrieve similar images, which outperforms state-of-art hand-crafted features. We propose an Open-Source-Intelligent (OSINT) real-time image search system, robust to retrieve modified images that allows first responders to analyze the current spread of images, sentiments floating and details of users propagating such content. The system also aids officials to save time of manually analyzing the content by reducing the search space on an average by 67 %.