Samir E. AbdelRahman

A machine Learning Approach for Opinion Holder Extraction in Arabic Language

Opinion mining aims at extracting useful subjective information from reliable amounts of text. Opinion mining holder recognition is a task that has not been considered yet in Arabic Language. This task essentially requires deep understanding of clauses structures. Unfortunately, the lack of a robust, publicly available, Arabic parser further complicates the research. This paper presents a leading research for the opinion holder extraction in Arabic news independent from any lexical parsers. We investigate constructing a comprehensive feature set to compensate the lack of parsing structural outcomes. The proposed feature set is tuned from English previous works coupled with our proposed semantic field and named entities features. Our feature analysis is based on Conditional Random Fields (CRF) and semi-supervised pattern recognition techniques. Different research models are evaluated via cross-validation experiments achieving 54.03 F-measure. We publicly release our own research outcome corpus and lexicon for opinion mining community to encourage further research.

Broad-coverage parsing using human-like memory constraints

Human syntactic processing shows many signs of taking place within a general-purpose short-term memory. But this kind of memory is known to have a severely constrained storage capacitypossibly constrained to as few as three or four distinct elements. This article describes a model of syntactic processing that operates successfully within these severe constraints, by recognizing constituents in a right-corner transformed representation (a variant of left-corner parsing) and mapping this representation to random variables in a Hierarchic Hidden Markov Model, a factored time-series model which probabilistically models the contents of a bounded memory store over time. Evaluations of the coverage of this model on a large syntactically annotated corpus of English sentences, and the accuracy of a a bounded-memory parsing strategy based on this model, suggest this model may be cognitively plausible.

Toward a Psycholinguistically-Motivated Model of Language Processing

Psycholinguistic studies suggest a model of human language processing that 1) performs incremental interpretation of spoken utterances or written text, 2) preserves ambiguity by maintaining competing analyses in parallel, and 3) operates within a severely constrained short-term memory store --- possibly constrained to as few as four distinct elements. This paper describes a relatively simple model of language as a factored statistical time-series process that meets all three of the above desiderata; and presents corpus evidence that this model is sufficient to parse naturally occurring sentences using human-like bounds on memory.

BluLab: Temporal Information Extraction for the 2015 Clinical TempEval Challenge

The 2015 Clinical TempEval Challenge addressed the problem of temporal reasoning in the clinical domain by providing an annotated corpus of pathology and clinical notes related to colon cancer patients. The challenge consisted of six subtasks: TIMEX3 and event span detection, TIMEX3 and event attribute classification, document relation time and narrative container relation classification. Our BluLab team participated in all six subtasks. For the TIMEX3 and event subtasks, we developed a ClearTK support vector machine pipeline using mainly simple lexical features along with information from rule-based systems. For the relation subtasks, we employed a conditional random fields classification approach, with input from a rule-based system for the narrative container relation subtask. Our team ranked first for all TIMEX3 and event subtasks, as well as for the document relation subtask.

Applying semantic-based probabilistic context-free grammar to medical language processing - A preliminary study on parsing medication sentences

Semantic-based sublanguage grammars have been shown to be an efficient method for medical language processing. However, given the complexity of the medical domain, parsers using such grammars inevitably encounter ambiguous sentences, which could be interpreted by different groups of production rules and consequently result in two or more parse trees. One possible solution, which has not been extensively explored previously, is to augment productions in medical sublanguage grammars with probabilities to resolve the ambiguity. In this study, we associated probabilities with production rules in a semantic-based grammar for medication findings and evaluated its performance on reducing parsing ambiguity. Using the existing data set from 2009 i2b2 NLP (Natural Language Processing) challenge for medication extraction, we developed a semantic-based CFG (Context Free Grammar) for parsing medication sentences and manually created a Treebank of 4564 medication sentences from discharge summaries. Using the Treebank, we derived a semantic-based PCFG (Probabilistic Context Free Grammar) for parsing medication sentences. Our evaluation using a 10-fold cross validation showed that the PCFG parser dramatically improved parsing performance when compared to the CFG parser.

A three-step approach for the derivation and validation of high-performing predictive models using an operational dataset: congestive heart failure readmission case study

BackgroundThe aim of this study was to propose an analytical approach to develop high-performing predictive models for congestive heart failure (CHF) readmission using an operational dataset with incomplete records and changing data over time.MethodsOur analytical approach involves three steps: pre-processing, systematic model development, and risk factor analysis. For pre-processing, variables that were absent in >50% of records were removed. Moreover, the dataset was divided into a validation dataset and derivation datasets which were separated into three temporal subsets based on changes to the data over time. For systematic model development, using the different temporal datasets and the remaining explanatory variables, the models were developed by combining the use of various (i) statistical analyses to explore the relationships between the validation and the derivation datasets; (ii) adjustment methods for handling missing values; (iii) classifiers; (iv) feature selection methods; and (iv) discretization methods. We then selected the best derivation dataset and the models with the highest predictive performance. For risk factor analysis, factors in the highest-performing predictive models were analyzed and ranked using (i) statistical analyses of the best derivation dataset, (ii) feature rankers, and (iii) a newly developed algorithm to categorize risk factors as being strong, regular, or weak.ResultsThe analysis dataset consisted of 2,787 CHF hospitalizations at University of Utah Health Care from January 2003 to June 2013. In this study, we used the complete-case analysis and mean-based imputation adjustment methods; the wrapper subset feature selection method; and four ranking strategies based on information gain, gain ratio, symmetrical uncertainty, and wrapper subset feature evaluators. The best-performing models resulted from the use of a complete-case analysis derivation dataset combined with the Class-Attribute Contingency Coefficient discretization method and a voting classifier which averaged the results of multi-nominal logistic regression and voting feature intervals classifiers. Of 42 final model risk factors, discharge disposition, discretized age, and indicators of anemia were the most significant. This model achieved a c-statistic of 86.8%.ConclusionThe proposed three-step analytical approach enhanced predictive model performance for CHF readmissions. It could potentially be leveraged to improve predictive model performance in other areas of clinical medicine.

A NEW EMAIL RETRIEVAL RANKING APPROACH

Email Retrieval task has recently taken much attention to help the user retrieve the email(s) related to the submitted query. Up to our knowledge, existing email retrieval ranking approaches sort the retrieved emails based on some heuristic rules, which are either search clues or some predefined user criteria rooted in email fields. Unfortunately, the user usually does not know the effective rule that acquires best ranking related to his query. This paper presents a new email retrieval ranking approach to tackle this problem. It ranks the retrieved emails based on a scoring function that depends on crucial email fields, namely subject, content, and sender. The paper also proposes an architecture to allow every user in a network/group of users to be able, if permissible, to know the most important network senders who are interested in his submitted query words. The experimental evaluation on Enron corpus prove that our approach outperforms known email retrieval ranking approaches..

An initial study of full parsing of clinical text using the Stanford Parser

Full parsing recognizes a sentence and generates a syntactic structure of it (a parse tree), which is useful for many natural language processing (NLP) applications. The Stanford Parser is one of the state-of-art parsers in the general English domain. However, there is no formal evaluation of its performance in clinical text that often contains ungrammatical structures. In this study, we randomly selected 50 sentences in the clinical corpus from 2010 i2b2 NLP challenge and manually annotated them to create a gold standard of parse trees. Our evaluation showed that the original Stanford Parser achieved a bracketing F-measure (BF) of 77% on the gold standard. Moreover, we assessed the effect of part-of-speech (POS) tags on parsing and our results showed that manually corrected POS tags achieved a maximum BF of 81%. Furthermore, we analyzed errors of the Stanford Parser and provided valuable insights to large-scale parse tree annotation for clinical text.

PREFCA: A portal retrieval engine based on formal concept analysis

Abstract The web is a network of linked sites whereby each site either forms a physical portal or a standalone page. In the former case, the portal presents an access point to its embedded web pages that coherently present a specific topic. In the latter case, there are millions of standalone web pages, that are scattered throughout the web, having the same topic and could be conceptually linked together to form virtual portals. Search engines have been developed to help users in reaching the adequate pages in an efficient and effective manner. All the known current search engine techniques rely on the web page as the basic atomic search unit. They ignore the conceptual links, that reveal the implicit web related meanings, among the retrieved pages. However, building a semantic model for the whole portal may contain more semantic information than a model of scattered individual pages. In addition, user queries can be poor and contain imprecise terms that do not reflect the real user intention. Consequently, retrieving the standalone individual pages that are directly related to the query may not satisfy the user’s need. In this paper, we propose PREFCA, a P ortal R etrieval E ngine based on F ormal C oncept A nalysis that relies on the portal as the main search unit. PREFCA consists of three phases: First, the information extraction phase that is concerned with extracting portal’s semantic data. Second, the formal concept analysis phase that utilizes formal concept analysis to discover the conceptual links among portal and attributes. Finally, the information retrieval phase where we propose a portal ranking method to retrieve ranked pairs of portals and embedded pages. Additionally, we apply the network analysis rules to output some portal characteristics. We evaluated PREFCA using two data sets, namely the Forum for Information Retrieval Evaluation 2010 and ClueWeb09 (category B) test data, for physical and virtual portals respectively. PREFCA proves higher F-measure accuracy, better Mean Average Precision ranking and comparable network analysis and efficiency results than other search engine approaches, namely Term Frequency Inverse Document Frequency (TF-IDF), Latent Semantic Analysis (LSA), and BM25 techniques. As well, it gains high Mean Average Precision in comparison with learning to rank techniques. Moreover, PREFCA also gains better reach time than Carrot as a well-known topic-based search engine.

Arabic Phrase-Level Contextual Polarity Recognition to Enhance Sentiment Arabic Lexical Semantic Database Generation

Most of opinion mining works need lexical resources for opinion which recognize the polarity of words (positive/ negative) regardless their contexts which called prior polarity. The word prior polarity may be changed when it is considered in its contexts, for example, positive words may be used in phrases expressing negative sentiments, or vice versa. In this paper, we aim at generating sentiment Arabic lexical semantic database having the word prior coupled with its contextual polarities and the related phrases. To do that, we study first the prior polarity effects of each word using our Sentiment Arabic Lexical Semantic Database on the sentence-level subjectivity and Support Vector Machine classifier. We then use the seminal English two-step contextual polarity phrase-level recognition approach to enhance word polarities within its contexts. Our results achieve significant improvement over baselines.