Abstract: This article describes our proposal for the Author Identification task in the PAN CLEF Challenge 2014. We adopt a machine learning approach based on several representations of the texts and on optimized decision trees which have as entry various attributes and which are learnt for every training corpus separately for this classification task. Our method ranked us at the 2nd place with an overall AUC of 70.7%, and C@1 of 68.4% and, between the 1st and the 6th place on the six corpora .

Abstract: This paper describes the Entrance Exams task at the CLEF QA Track 2014. Following 2013 edition, the data set has been extracted from actual uni- versity entrance examinations including a variety of topics and question types. Systems receive a set of Multiple-Choice Reading Comprehension tests where the task is to select the correct answer among a finite set of candidates, accord- ing to the given text. Questions are designed originally for testing human exam- inees, rather than evaluating computer systems. Therefore, the data set chal- lenges human ability to show their understanding of texts. Thus, questions and answers are lexically distant from their supporting excerpts in text, requiring not only a high degree of textual inference, but also the development of strate- gies for selecting the correct answer. As a novelty this year, data sets originally in English were manually translated into Russian, French, Spanish and Italian.

Abstract: This paper reports the outcomes of a longitudinal study on the CLEF Ad Hoc track in order to assess its impact on the effectiveness of monolingual, bilingual and multilingual information access and retrieval systems. Monolingual retrieval shows a positive trend, even if the performance increase is not always steady from year to year; bilingual retrieval has demonstrated higher improvements in recent years, probably due to the better linguistic resources now available; and, multilingual retrieval exhibits constant improvement and performances comparable to bilingual (and, sometimes, even monolingual) ones.

Abstract: According to the organizers, the objective of the 2014 CLEF- INEX Tweet Contextualization Task is: \...The Tweet Contextualization aims at providing automatically information - a summary that explains the tweet. This requires combining multiple types of processing from in- formation retrieval to multi-document summarization including entity linking." We present three statistical summarizer systems applied to the CLEF-INEX 2014 task. Cortex summarizer uses several sentence selec- tion metrics and an optimal decision module to score sentences from a document source. Artex summarizer uses a simple inner product among the topic-vector and the pseudo-word vector. Reg summarizer is a per- formant graph-based summarizer. The results show that our systems performed well on CLEF-INEX task. Our three systems have obtained the rst rank in the INEX manual evaluation.

Abstract: This paper describes the methods we have employed to solve the au- thor profiling task at PAN-2014. Our goal was to rely mainly on features from Information Retrieval to identify the age group and the gender of the author of a given text. We describe the features, the classification algorithms employed, and how the experiments were run. Also, we provide an analysis of our results compared to other groups.

Abstract: This paper presents our first participation in user-centred health information retrieval task at the CLEFeHealth 2014. This task has as objective the information retrieval to answer patients’ questions when reading clinical reports. We have submitted only the mandatory run (baseline system). The obtained results are motivating with map=0.1677 and p@10=0.5460 but can be improved.

Abstract: This paper describes the CLEF QA Track 2014. In the current general scenario for the CLEF QA Track, the starting point is always a natural language question. However, answering some questions may need to query Linked Data (especially if aggregations or logical inferences are required), some questions may need textual inferences and querying free text, and finally, answering some queries may require both sources of information. The track was divided into three tasks: QALD focused on translating natural language questions into SPARQL queries; BioASQ focused on the biomedical domain, and Entrance Exams focused on answering questions to assess machine reading capabilities.

Abstract: In this paper we, Team IRLabDAIICT, describe our participation in the ShARe/CLEF ehealth 2014 task 3: Information Retrieval for addressing questions related to patients health based on clinical reports. We submitted a total of six runs out of the seven in this years task. In our approach we focus on examining the relevance between the documents and user generated query by conducting experiments through query analysis. Our major challenge is to bridge the conceptual gap between the user-generated queries (in-formal query) to biomedical specific terminology (formal query). We incorporate the MeSH (Medical Subject Headings) library , which is a medical thesaurus mapping layman terms to medical synonym terms in order to target the concept matching problem. We use blind relevance feedback model for relevance feedback and query-likelihood model for query expansion which performed the best in the experiments conducted by us. The retrieval system is evaluated based on various parameters as: mean average precision, precision (P@5), precision (P@10), NDCG@5 and NDCG@10, with P@10 and NDCG@10 being the primary and secondary evaluation measures. The experiments were conducted on the gigantic 43.6 GB ShARe/CLEF 2013 Task 3 dataset harvested using (a) EU-FP7 Khresmoi project and and (b) a new 2014 set of English general realistic public queries based on the discharge summary contents. We have obtained the highest result in our baseline run (run 1), with compared to our other five runs, which is 0.706 as declared by ShARe/CLEF organizing committee. We further propose to incorporate a machine learning based retrieval algorithm prediction model for further exploration.

Abstract: This article describes our proposal for the Author Identification task in the PAN CLEF Challenge 2014. We have adopted a machine learning ap- proach based on several representations of the texts and on optimized decision trees which have as entry various attributes and which are learned for every train- ing corpus separately for this classification task. Our method ranked us at the 2nd place with an overall AUC of 70.7%, and C@1 of 68.4% and, between the 1st and the 6th place on the six corpora.

Abstract: This paper describes the participation of the SNUMedinfo team at the BioASQ Task 2a and Task 2b of CLEF 2014 Question Answering track. Task 2a was about biomedical semantic indexing. We trained SVM classifiers to auto- matically assign relevant MeSH descriptors to the MEDLINE article. Regarding Task 2b biomedical question answering, we participated at the document retrieval subtask in Phase A and the ideal answer generation subtask in Phase B. In the document retrieval task, we mostly experimented with semantic concept-en- riched dependence model and sequential dependence model. Semantic concept- enriched dependence model showed significant improvement over baseline. In the ideal answer generation task, we reformulated task as, given relevant lists of passages, selecting the best ones to build the answer. We applied three heuristic methods.

Abstract: This book constitutes the refereed proceedings of the 5th International Conference of the CLEF Initiative, CLEF 2014, held in Sheffield, UK, in September 2014. The 11 full papers and 5 short papers presented were carefully reviewed and selected from 30 submissions. They cover a broad range of issues in the fields of multilingual and multimodal information access evaluation, also included are a set of labs and workshops designed to test different aspects of mono and cross-language information retrieval systems

Abstract: In this paper, the basics of the three tuning approaches of the evolving CoReMo Plagiarism Detector are shown, focused for the Text Alignment task. In the last PAN edition, it was observed that the different corpora could condition the necessary tuning, and the results using an overfitted tuning from a different corpus could be far from the expected ones. This year's goal has been to find the way to get the system could be self­tuned, looking for improving the performance of any fixed parameter tuning, and to be very closed to the over­fitted performance for any corpus. All of these tuning approaches have a high Plagdet performance for any corpus, but it’s intended to show the different advances effect on each corpus and for all the years corpora. They include new features for parameters self­tuning, based on the size and the ratio of the compared documents. For the competition, our choice was based on the most constant detection quality tuning approach when any condition (called WideTuning).

Abstract: The task 3 of CLEF eHealth Evaluation lab aims to help laypeople get more accurate information from health related documents. In this task, we did several experiments and tried dierent technolo- gies to improve the retrieval performance. We tried to clean the original dataset and did sentence level retrieval. We explored dierent parameter settings for pseudo relevance feedback. Description and Narrative was utilized to expand the query as well. We also modied Markov Random Field (MRF) model to expand the query using medical phrase only. In our training set (2013 test set), using those methods can signicantly im- prove the retrieval performance by 8-15% from baseline. We submitted 4 runs. Results on 2014 test set suggest that the technologies we used except MRF have the potential to improve the performance for the top 5 retrieved results.

Abstract: This paper describes the participation of RePaLi, a team composed with members of IRISA, LIMSI and STL, to the biomedical information retrieval challenge proposed in the framework of CLEF eHealth. For this first participation, our approach relies on a state-of-the-art IR system called Indri, based on statistical language modeling, and on semantic resources. The purpose of semantic resources and methods is to manage the term variation such as synonyms, morpho-syntactic variants, abbreviation or nested terms. Different combinations of resources and Indri settings are explored, mostly based on query expansion. For the runs submitted, our system shows up to 67.40 p@10 and up to 67.93 NDCG@10.

Abstract: Encouraged by results from our approaches in previous PAN workshops, this paper explores three different approaches using stopword cooccurrence. High frequency patterns of co-occurrence can be used to some extent as identifiers of an author’s style, and have been demonstrated to operate similarly across certain languages - without requiring deeper linguistic knowledge. However, making best use of such information remains unresolved. We compare results from applying three approaches overs such patterns: a frequency-mean-variance framework; a positional-frequency cosine comparison approach, and a cosine distance-based approach. A clearly advantageous approach across all languages and genres is yet to emerge.

Abstract: This paper presents the details of participation of DEMIR (Dokuz Eylül University Multimedia Information Retrieval) research team to the Share/CLEF eHealth 2014. This year, we participated to task 3a: monolingual user-centered health information retrieval. In this task, we focused to apply query expansion techniques selectively to some queries to improve the performance of information retrieval. Thus, we first extracted some statistical features from queries such as length of query, sum and intersect of document frequencies of each query term etc. We develop a system to predict if a query is to be expanded or not. Then, we trained our system with previous year’s data. Then, we applied a query expansion method only to the queries, which are selected by the system. The results show that the approach we proposed slightly improves our baseline retrieval performance in terms of P@10.

Abstract: BiTeM/SIBtex is a university research group with a strong background in Text Mining and Bibliomics, and a long tradition of participating in large evaluation campaigns. The CLEF RepLab 2014 Track was the occasion to integrate several local tools into a complete system for tweet monitoring and categorization based on instance-based learning. The algorithm we implemented was a k Nearest Neighbors. Dealing with the domain (automotive or banking) and the language (English or Spanish), the experiments showed that the categorizer was not affected by the choice of representation: even with all data merged into one single Knowledge Base (KB), the observed performances were close to those with dedicated KBs. Furthermore, English training data in addition to the sparse Spanish data were useful for Spanish categorization (+14% for accuracy for automotive, +26% for banking). Finally, our best official run was in top five. Yet, performances suffered from an overprediction of the most prevalent category, while we were not able to address this issue of unbalanced labels within the competition time. The algorithm showed the defects of its virtues: it was very robust, but not easy to improve. BiTeM/SIBtex tools for tweet monitoring are available within the DrugsListener Project page of the BiTeM website (http://bitem.hesge.ch/).

Abstract: This paper describes a method that was implemented in the software submitted to PAN 2014 competition for the source retrieval task. For generating queries we use the most important noun phrases and words of sentences selected from a given suspicious document. To download documents that are likely to be sources of plagiarism we employ a sentence similarity measure.

Abstract: This paper describes the CLAS system which accepts natural language queries in the domain of music theory to p erform passage retrieval from a musical score. This syste m was produced for participation in the C@merata MediaEval 2014 shared task. The system uses a domain-specific parser to interpr et the query and answer generation methods based on feature unification. Performance on this task was encouraging with 0.76 precision and 0.96 recall. 1. INTRODUCTION This paper describes the CLAS system which selects processes and retrieves potentially relevant answer s from structured data given a natural language query. In this work, the queries and the structured data are in the domain o f music theory, as defined by the C@merata 2014 task (1). The CLAS system produces candidate answers by selecting passages fr om an musical score (in XML). Answers may be any consecutive time points spanning multiple whole and partial bars. For example, a query ``4 crotchets'' should retriev e any sequence of four consecutive elements in the score where each element is a note and each note has the time durati on of a crotchet (one quarter of a whole note). In such a system, e xpert knowledge is needed to interpret the query. However, this no t just limited to definitions of musical concepts (e.g., ``crotchet'' ). For example, the query ``4 crotchets'' should be interpreted not just as any four notes with crotchet duration within the music (comp are this to a general knowledge query ``4 composers'' requiring a ny four musical composers to be provided) but specifically four notes in sequence. Furthermore, these four notes would typic ally be expected to be in the same voice or part ; for example, if it were a piano score for two hands, the four crotchets might be a sequence written in the treble clef, played by the right han d. In this paper, we describe a system that processes the input query, mapping from words in English to music metadata corresponding to the search criteria, or features, represented as a set of attribute-value pairs. An exhaustive search of an XML score is performed, note by note, for candidate ans wers using feature unification. This system achieved an overall performance of 0.76 precision and 0.96 recall. The remainder of the pa per outlines the system in more detail and presents the C@merata evaluation results.

Abstract: This paper describes the participation of LABERINTO team at the ShARe/CLEF eHealth Evaluation Lab 2014 task 3a. We perform four di erent experiments which consist of a baseline and three variants of the baseline model. The rst was mandatory baseline system with only title and description in the query. Our baseline retrieval system used a Lucene Index scheme with traditional stopping and stemming, no external resources was used. We submitted three additional runs (without the discharge summaries), two from a Lucene-based system with MeSH query expansion and one of which made use of the National Library of Medicine's MetaMap tool to perform term boosting.

Abstract: In this article, I argue for a reconsideration of the roman a clef, using Caroline Lamb’s Glenarvon (1816) as a case study. I challenge the conventional narrative of the rise of the novel that labels the roman a clef as a vestigial form with no place in relation to the dominant realist novel. In making this argument, I establish a social reading practice, singular to romans a clef, using archival materials to illustrate the ways in which Regency readers circulated keys and gossip about Glenarvon and Lamb’s life as a part of their response to the novel. Such a reading practice opens up possibilities for a marginalized writer, especially a marginalized female writer like Lamb, to disseminate gossip about herself as a way to enter a literary marketplace.

Abstract: This paper describes our participation to the Entrance Ex- ams Task of CLEF 2014's Question Answering Track. The goal is to an- swer multiple-choice questions on short texts. Our system rst retrieves passages relevant to the question, through lexical expansion involving a structured use of the Simple English Wiktionary and WordNet. Then it extracts predicate-argument structures (PAS) from each answer choice and aligns them to PAS found in the passages retrieved in the rst step. Finally, manually crafted rules are applied to those alignments to try to invalidate answer choices. If enough answer choices are thus invalidated, we make a decision on the remaining answer choices based on their align- ment scores with the passages. We submitted several runs in the task, only one of which reached the random baseline (c@1 of 0.25). In the last section, we provide an analysis of the dierences between our relatively good results obtained on trial data and the poor performance of our test run.

Abstract: This paper describes a learning system developed for the RepLab 2014 author profiling task at UNED. The system uses a voting model, which employs a small set of features based mainly on the tweet text information such as POS tags, number of hashtags or number of links. In the unofficial run, the feature set was increased with Twitter metadata such as number of followers or retweet speed. The system achieved good results in author categorisation, although its performance in author ranking was low.

Abstract: This study was conducted to assess physical and perceptual properties of a tactile display for a vibrotactile notification system within the CIRMMT Live Electronics Framework (CLEF), a Max-based mo ...

Abstract: We present our rst participation in a ShARe/CLEF eHealth Lab contributing for task 2a. Task 2 is an extension of the 2013 lab task 1 and consists of information extraction from clinical texts for Disease/Disorder Template Filling; task 2a aims at predicting each at- tribute's normalization value. This work constitutes a preliminary approach to the problem of extract- ing and handling information from clinical texts. More than getting a good result, our priority was to get a rst hint on the questions and problems that are posed within this area. For that, we developed a system that combines information from cTAKES output and the training corpus. The performance was measured using ac- curacy. Our system ranked 7th with an accuracy of 0.802, a F1 of 0.214, a precision of 0.217 and a recall value of 0.211.

Abstract: Disease/Disorder Semantic Template Filling – Information Extraction Challenge in the ShARe/CLEF eHealth Evaluation Lab 2014

Abstract: With the continuously growing literatures in the biomedical domain, it is not feasible for researchers to manually go through all information for answering questions The task of making knowledge contained in texts in forms that machines can use for automated processing is more and more important This paper describes a system to answer multiple-choice questions for the biomedical domain while reading a given document In this study, we use the data from the pilot task "machine reading of biomedical texts about Alzheimer's disease" which is a task of the Question Answering for Machine Reading Evaluation QA4MRE Lab at CLEF 2012 We adapt the concept of answer validation that assumes the over-generation hypotheses will be checked in the validation step In the following, the query expansion technique "global analysis" is applied The best result is 051 c@1 score which is clearly above the baseline at CLEF 2012 and shows an exhilarating performance