Linear Conditional Random Field Research Articles

A stratified approach to function fingerprinting in program binaries using diverse features

Fingerprinting individual functions in binary code is useful in many security applications ranging from digital forensic analysis of malware corpora to the detection of critical security vulnerabilities. However, existing approaches for fingerprinting functions are typically not resilient to code transformation methods or the use of different compilers. Moreover, another common weakness with these approaches is that when they report a similarity, they do not provide reverse engineers with any insight into the underlying evidence. In order to bridge this gap, our paper presents Plumeria, an obfuscation-resilient and scalable approach based on a stratified architecture comprised of three layers. The first layer retrieves as many candidates as possible by capturing statistical characteristics, function behavior, and function neighborhood relationships. The second layer then trains a linear conditional random field to learn the correlations between the features of the function and its semantics. This layer is designed to reduce the number of false positives. Finally, the third layer is designed to provide insights into the underlying evidence by collecting the side effects exhibited from the candidates selected by the previous layer. Our study evaluates Plumeria in the context of several scenarios: fingerprinting functions in obfuscated/de-obfuscated binaries; fingerprinting functions across different compilers; fingerprinting various vulnerabilities across compilers and versions; and fingerprinting standard library functions. We then benchmark Plumeria on real-world projects and malware binaries, comparing it with existing state-of-the-art solutions. Our results show that Plumeria outperforms existing solutions, with an average precision of over 89%

Guiding Attention in Sequence-to-Sequence Models for Dialogue Act Prediction

The task of predicting dialog acts (DA) based on conversational dialog is a key component in the development of conversational agents. Accurately predicting DAs requires a precise modeling of both the conversation and the global tag dependencies. We leverage seq2seq approaches widely adopted in Neural Machine Translation (NMT) to improve the modelling of tag sequentiality. Seq2seq models are known to learn complex global dependencies while currently proposed approaches using linear conditional random fields (CRF) only model local tag dependencies. In this work, we introduce a seq2seq model tailored for DA classification using: a hierarchical encoder, a novel guided attention mechanism and beam search applied to both training and inference. Compared to the state of the art our model does not require handcrafted features and is trained end-to-end. Furthermore, the proposed approach achieves an unmatched accuracy score of 85% on SwDA, and state-of-the-art accuracy score of 91.6% on MRDA.

CheNER: a tool for the identification of chemical entities and their classes in biomedical literature.

BackgroundSmall chemical molecules regulate biological processes at the molecular level. Those molecules are often involved in causing or treating pathological states. Automatically identifying such molecules in biomedical text is difficult due to both, the diverse morphology of chemical names and the alternative types of nomenclature that are simultaneously used to describe them. To address these issues, the last BioCreAtIvE challenge proposed a CHEMDNER task, which is a Named Entity Recognition (NER) challenge that aims at labelling different types of chemical names in biomedical text.MethodsTo address this challenge we tested various approaches to recognizing chemical entities in biomedical documents. These approaches range from linear Conditional Random Fields (CRFs) to a combination of CRFs with regular expression and dictionary matching, followed by a post-processing step to tag those chemical names in a corpus of Medline abstracts. We named our best performing systems CheNER.ResultsWe evaluate the performance of the various approaches using the F-score statistics. Higher F-scores indicate better performance. The highest F-score we obtain in identifying unique chemical entities is 72.88%. The highest F-score we obtain in identifying all chemical entities is 73.07%. We also evaluate the F-Score of combining our system with ChemSpot, and find an increase from 72.88% to 73.83%.ConclusionsCheNER presents a valid alternative for automated annotation of chemical entities in biomedical documents. In addition, CheNER may be used to derive new features to train newer methods for tagging chemical entities. CheNER can be downloaded from http://metres.udl.cat and included in text annotation pipelines.

Semantic-Based Requirements Content Management for Cloud Software

Cloud Software is a software complex system whose topology and behavior can evolve dynamically in Cloud-computing environments. Given the unpredictable, dynamic, elasticity, and on-demand nature of the Cloud, it would be unrealistic to assume that traditional software engineering can “cleanly” satisfy the behavioral requirements of Cloud Software. In particular, the majority of traditional requirements managements take document-centric approaches, which have low degree of automation, coarse-grained management, and limited support for requirements modeling activities. Facing the challenges, based on metamodeling frame called RGPS (Role-Goal-Process-Service) international standard, this paper firstly presents a hierarchical framework of semantic-based requirements content management for Cloud Software. And then, it focuses on some of the important management techniques in this framework, such as the native storage scheme, an ordered index with keywords, requirements instances classification based linear conditional random fields (CRFs), and breadth-first search algorithm for associated instances. Finally, a prototype tool called RGPS-RM for semantic-based requirements content management is implemented to provide supporting services for open requirements process of Cloud Software. The proposed framework applied to the Cloud Software development is demonstrated to show the validity and applicability. RGPS-RM also displays effect of fine-grained retrieval and breadth-first search algorithm for associated instance in visualization.

Named entity recognition for tweets

Two main challenges of Named Entity Recognition (NER) for tweets are the insufficient information in a tweet and the lack of training data. We propose a novel method consisting of three core elements: (1) normalization of tweets; (2) combination of a K-Nearest Neighbors (KNN) classifier with a linear Conditional Random Fields (CRF) model; and (3) semisupervised learning framework. The tweet normalization preprocessing corrects common ill-formed words using a global linear model. The KNN-based classifier conducts prelabeling to collect global coarse evidence across tweets while the CRF model conducts sequential labeling to capture fine-grained information encoded in a tweet. The semisupervised learning plus the gazetteers alleviate the lack of training data. Extensive experiments show the advantages of our method over the baselines as well as the effectiveness of normalization, KNN, and semisupervised learning.

Detection and characterization of regulatory elements using probabilistic conditional random field and hidden Markov models

By altering the electrostatic charge of histones or providing binding sites to protein recognition molecules, Chromatin marks have been proposed to regulate gene expression, a property that has motivated researchers to link these marks to Cis-regulatory elements. With the help of next generation sequencing technologies, we can now correlate one specific chromatin mark with regulatory elements (e.g. enhancers or promoters) and also build tools, such as hidden Markov models, to gain insight into mark combinations. However, hidden Markov models have limitation for their character of generative models and assume that a current observation depends only on a current hidden state in the chain. Here, we employed two graphical probabilistic models, namely the linear conditional random field model and multivariate hidden Markov model, to mark gene regions with different states based on recurrent and spatially coherent character of these eight marks. Both models revealed chromatin states that may correspond to enhancers and promoters, transcribed regions, transcriptional elongation, and low-signal regions. We also found that the linear conditional random field model was more effective than the hidden Markov model in recognizing regulatory elements, such as promoter-, enhancer-, and transcriptional elongation-associated regions, which gives us a better choice.

Two-stage NER for tweets with clustering

One main challenge of Named Entities Recognition (NER) for tweets is the insufficient information in a single tweet, owing to the noisy and short nature of tweets. We propose a novel system to tackle this challenge, which leverages redundancy in tweets by conducting two-stage NER for multiple similar tweets. Particularly, it first pre-labels each tweet using a sequential labeler based on the linear Conditional Random Fields (CRFs) model. Then it clusters tweets to put tweets with similar content into the same group. Finally, for each cluster it refines the labels of each tweet using an enhanced CRF model that incorporates the cluster level information, i.e., the labels of the current word and its neighboring words across all tweets in the cluster. We evaluate our method on a manually annotated dataset, and show that our method boosts the F1 of the baseline without collectively labeling from 75.4% to 82.5%.

Complex Terminology Extraction Model from Unstructured Web Text Based Linguistic and Statistical Knowledge

Textual data remain the most interesting source of information in the web. In the authors’ research, they focus on a very specific kind of information namely “complex terms”. Indeed, complex terms are defined as semantic units composed of several lexical units that can describe in a relevant and exhaustive way the text content. In this paper, they present a new model for complex terminology extraction (COTEM), which integrates linguistic and statistical knowledge. Thus, the authors try to focus on three main contributions: firstly, they show the possibility of using a linear Conditional Random Fields (CRF) for complex terminology extraction from a specialized text corpus. Secondly, prove the ability of a Conditional Random Field to model linguistic knowledge by incorporating grammatical observations in the CRF’s features. Finally, the authors present the benefits gained by the integration of statistical knowledge on the quality of the terminology extraction.