An extensible framework for analysing resource behaviour using event logs

Business processes depend on human resources and managers must regularly evaluate the performance of their employees based on a number of measures, some of which are subjective in nature. As modern organisations use information systems to automate their business processes and record information about processes’ executions in event logs, it now becomes possible to get objective information about resource behaviours by analysing data recorded in event logs. We present an extensible framework for extracting knowledge from event logs about the behaviour of a human resource and for analysing the dynamics of this behaviour over time. The framework is fully automated and implements a predefined set of behavioural indicators for human resources. It also provides a means for organisations to define their own behavioural indicators, using the conventional Structured Query Language, and a means to analyse the dynamics of these indicators. The framework's applicability is demonstrated using an event log from a German bank.

Learning detectors quickly with stationary statistics

Computer vision is increasingly becoming interested in the rapid estimation of object detectors. The canonical strategy of using Hard Negative Mining to train a Support Vector Machine is slow, since the large negative set must be traversed at least once per detector. Recent work has demonstrated that, with an assumption of signal stationarity, Linear Discriminant Analysis is able to learn comparable detectors without ever revisiting the negative set. Even with this insight, the time to learn a detector can still be on the order of minutes. Correlation filters, on the other hand, can produce a detector in under a second. However, this involves the unnatural assumption that the statistics are periodic, and requires the negative set to be re-sampled per detector size. These two methods differ chie y in the structure which they impose on the co- variance matrix of all examples. This paper is a comparative study which develops techniques (i) to assume periodic statistics without needing to revisit the negative set and (ii) to accelerate the estimation of detectors with aperiodic statistics. It is experimentally verified that periodicity is detrimental.

Key-private proxy re-encryption under LWE

Proxy re-encryption (PRE) is a highly useful cryptographic primitive whereby Alice and Bob can endow a proxy with the capacity to change ciphertext recipients from Alice to Bob, without the proxy itself being able to decrypt, thereby providing delegation of decryption authority. Key-private PRE (KP-PRE) specifies an additional level of confidentiality, requiring pseudo-random proxy keys that leak no information on the identity of the delegators and delegatees. In this paper, we propose a CPA-secure PK-PRE scheme in the standard model (which we then transform into a CCA-secure scheme in the random oracle model). Both schemes enjoy highly desirable properties such as uni-directionality and multi-hop delegation. Unlike (the few) prior constructions of PRE and KP-PRE that typically rely on bilinear maps under ad hoc assumptions, security of our construction is based on the hardness of the standard Learning-With-Errors (LWE) problem, itself reducible from worst-case lattice hard problems that are conjectured immune to quantum cryptanalysis, or “post-quantum”. Of independent interest, we further examine the practical hardness of the LWE assumption, using Kannan’s exhaustive search algorithm coupling with pruning techniques. This leads to state-of-the-art parameters not only for our scheme, but also for a number of other primitives based on LWE published the literature.

Mining specific features for acquiring user information needs

Term-based approaches can extract many features in text documents, but most include noise. Many popular text-mining strategies have been adapted to reduce noisy information from extracted features; however, text-mining techniques suffer from low frequency. The key issue is how to discover relevance features in text documents to fulfil user information needs. To address this issue, we propose a new method to extract specific features from user relevance feedback. The proposed approach includes two stages. The first stage extracts topics (or patterns) from text documents to focus on interesting topics. In the second stage, topics are deployed to lower level terms to address the low-frequency problem and find specific terms. The specific terms are determined based on their appearances in relevance feedback and their distribution in topics or high-level patterns. We test our proposed method with extensive experiments in the Reuters Corpus Volume 1 dataset and TREC topics. Results show that our proposed approach significantly outperforms the state-of-the-art models.

Optimization of an integrated model for automatic reduction and expansion of long queries

A long query provides more useful hints for searching relevant documents, but it is likely to introduce noise which affects retrieval performance. In order to smooth such adverse effect, it is important to reduce noisy terms, introduce and boost additional relevant terms. This paper presents a comprehensive framework, called Aspect Hidden Markov Model (AHMM), which integrates query reduction and expansion, for retrieval with long queries. It optimizes the probability distribution of query terms by utilizing intra-query term dependencies as well as the relationships between query terms and words observed in relevance feedback documents. Empirical evaluation on three large-scale TREC collections demonstrates that our approach, which is automatic, achieves salient improvements over various strong baselines, and also reaches a comparable performance to a state of the art method based on user’s interactive query term reduction and expansion.

Functional encryption for threshold functions (or fuzzy IBE) from lattices

Cryptosystems based on the hardness of lattice problems have recently acquired much importance due to their average-case to worst-case equivalence, their conjectured resistance to quantum cryptanalysis, their ease of implementation and increasing practicality, and, lately, their promising potential as a platform for constructing advanced functionalities. In this work, we construct “Fuzzy” Identity Based Encryption from the hardness of the Learning With Errors (LWE) problem. We note that for our parameters, the underlying lattice problems (such as gapSVP or SIVP) are assumed to be hard to approximate within supexponential factors for adversaries running in subexponential time. We give CPA and CCA secure variants of our construction, for small and large universes of attributes. All our constructions are secure against selective-identity attacks in the standard model. Our construction is made possible by observing certain special properties that secret sharing schemes need to satisfy in order to be useful for Fuzzy IBE. We also discuss some obstacles towards realizing lattice-based attribute-based encryption (ABE).

Bitter to better - how to make bitcoin a better currency

Bitcoin is a distributed digital currency which has attracted a substantial number of users. We perform an in-depth investigation to understand what made Bitcoin so successful, while decades of research on cryptographic e-cash has not lead to a large-scale deployment. We ask also how Bitcoin could become a good candidate for a long-lived stable currency. In doing so, we identify several issues and attacks of Bitcoin, and propose suitable techniques to address them.

Kamouflage : loss-resistant password management

We introduce Kamouflage: a new architecture for building theft-resistant password managers. An attacker who steals a laptop or cell phone with a Kamouflage-based password manager is forced to carry out a considerable amount of online work before obtaining any user credentials. We implemented our proposal as a replacement for the built-in Firefox password manager, and provide performance measurements and the results from experiments with large real-world password sets to evaluate the feasibility and effectiveness of our approach. Kamouflage is well suited to become a standard architecture for password managers on mobile devices.

Shrinking the keys of discrete-log-type lossy trapdoor functions

To this day, realizations in the standard-model of (lossy) trapdoor functions from discrete-log-type assumptions require large public key sizes, e.g., about Θ(λ 2) group elements for a reduction from the decisional Diffie-Hellman assumption (where λ is a security parameter). We propose two realizations of lossy trapdoor functions that achieve public key size of only Θ(λ) group elements in bilinear groups, with a reduction from the decisional Bilinear Diffie-Hellman assumption. Our first construction achieves this result at the expense of a long common reference string of Θ(λ 2) elements, albeit reusable in multiple LTDF instantiations. Our second scheme also achieves public keys of size Θ(λ), entirely in the standard model and in particular without any reference string, at the cost of a slightly more involved construction. The main technical novelty, developed for the second scheme, is a compact encoding technique for generating compressed representations of certain sequences of group elements for the public parameters.

Lattice mixing and vanishing trapdoors : a framework for fully secure short signatures and more

We propose a framework for adaptive security from hard random lattices in the standard model. Our approach borrows from the recent Agrawal-Boneh-Boyen families of lattices, which can admit reliable and punctured trapdoors, respectively used in reality and in simulation. We extend this idea to make the simulation trapdoors cancel not for a specific forgery but on a non-negligible subset of the possible challenges. Conceptually, we build a compactly representable, large family of input-dependent “mixture” lattices, set up with trapdoors that “vanish” for a secret subset which we hope the forger will target. Technically, we tweak the lattice structure to achieve “naturally nice” distributions for arbitrary choices of subset size. The framework is very general. Here we obtain fully secure signatures, and also IBE, that are compact, simple, and elegant.

Strong cryptography from weak secrets

Distributed-password public-key cryptography (DPwPKC) allows the members of a group of people, each one holding a small secret password only, to help a leader to perform the private operation, associated to a public-key cryptosystem. Abdalla et al. recently defined this tool [1], with a practical construction. Unfortunately, the latter applied to the ElGamal decryption only, and relied on the DDH assumption, excluding any recent pairing-based cryptosystems. In this paper, we extend their techniques to support, and exploit, pairing-based properties: we take advantage of pairing-friendly groups to obtain efficient (simulation-sound) zero-knowledge proofs, whose security relies on the Decisional Linear assumption. As a consequence, we provide efficient protocols, secure in the standard model, for ElGamal decryption as in [1], but also for Linear decryption, as well as extraction of several identity-based cryptosystems [6,4]. Furthermore, we strenghten their security model by suppressing the useless testPwd queries in the functionality.

HPAKE : Password authentication secure against cross-site user impersonation

We propose a new kind of asymmetric mutual authentication from passwords with stronger privacy against malicious servers, lest they be tempted to engage in “cross-site user impersonation” to each other. It enables a person to authenticate (with) arbitrarily many independent servers, over adversarial channels, using a memorable and reusable single short password. Beside the usual PAKE security guarantees, our framework goes to lengths to secure the password against brute-force cracking from privileged server information.

Distributed public-key cryptography from weak secrets

We introduce the notion of distributed password-based public-key cryptography, where a virtual high-entropy private key is implicitly defined as a concatenation of low-entropy passwords held in separate locations. The users can jointly perform private-key operations by exchanging messages over an arbitrary channel, based on their respective passwords, without ever sharing their passwords or reconstituting the key. Focusing on the case of ElGamal encryption as an example, we start by formally defining ideal functionalities for distributed public-key generation and virtual private-key computation in the UC model. We then construct efficient protocols that securely realize them in either the RO model (for efficiency) or the CRS model (for elegance). We conclude by showing that our distributed protocols generalize to a broad class of “discrete-log”-based public-key cryptosystems, which notably includes identity-based encryption. This opens the door to a powerful extension of IBE with a virtual PKG made of a group of people, each one memorizing a small portion of the master key.

The uber-assumption family

We offer an exposition of Boneh, Boyen, and Goh’s “uber-assumption” family for analyzing the validity and strength of pairing assumptions in the generic-group model, and augment the original BBG framework with a few simple but useful extensions.

Expressive subgroup signatures

In this work, we propose a new generalization of the notion of group signatures, that allows signers to cover the entire spectrum from complete disclosure to complete anonymity. Previous group signature constructions did not provide any disclosure capability, or at best a very limited one (such as subset membership). Our scheme offers a very powerful language for disclosing exactly in what capacity a subgroup of signers is making a signature on behalf of the group.