Ebaluatoia: crowd evaluation of English-Basque machine translation

[EU]Lan honetan Ebaluatoia aurkezten da, eskala handiko ingelesa-euskara itzulpen automatikoko ebaluazio kanpaina, komunitate-elkarlanean oinarritua. Bost sistemaren itzulpen kalitatea konparatzea izan da kanpainaren helburua, zehazki, bi sistema estatistiko, erregeletan oinarritutako bat eta sistema hibrido bat (IXA taldean garatuak) eta Google Translate. Emaitzetan oinarrituta, sistemen sailkapen bat egin dugu, baita etorkizuneko ikerkuntza bideratuko duten zenbait analisi kualitatibo ere, hain zuzen, ebaluazio-bildumako azpi-multzoen analisia, iturburuko esaldien analisi estrukturala eta itzulpenen errore-analisia. Lanak analisi hauen hastapenak aurkezten ditu, etorkizunean zein motatako analisietan sakondu erakutsiko digutenak.

Distributional semantics and machine learning for statistical machine translation

[EU]Lan honetan semantika distribuzionalaren eta ikasketa automatikoaren erabilera aztertzen dugu itzulpen automatiko estatistikoa hobetzeko. Bide horretan, erregresio logistikoan oinarritutako ikasketa automatikoko eredu bat proposatzen dugu hitz-segiden itzulpen- probabilitatea modu dinamikoan modelatzeko. Proposatutako eredua itzulpen automatiko estatistikoko ohiko itzulpen-probabilitateen orokortze bat dela frogatzen dugu, eta testuinguruko nahiz semantika distribuzionaleko informazioa barneratzeko baliatu ezaugarri lexiko, hitz-cluster eta hitzen errepresentazio bektorialen bidez. Horretaz gain, semantika distribuzionaleko ezagutza itzulpen automatiko estatistikoan txertatzeko beste hurbilpen bat lantzen dugu: hitzen errepresentazio bektorial elebidunak erabiltzea hitz-segiden itzulpenen antzekotasuna modelatzeko. Gure esperimentuek proposatutako ereduen baliagarritasuna erakusten dute, emaitza itxaropentsuak eskuratuz oinarrizko sistema sendo baten gainean. Era berean, gure lanak ekarpen garrantzitsuak egiten ditu errepresentazio bektorialen mapaketa elebidunei eta hitzen errepresentazio bektorialetan oinarritutako hitz-segiden antzekotasun neurriei dagokienean, itzulpen automatikoaz haratago balio propio bat dutenak semantika distribuzionalaren arloan.

EUSMT: incorporating linguistic information to SMT for a morphologically rich language. Its use in SMT-RBMT-EBMT hybridation

148 p.: graf.

Exploring Semantic Textual Similarity

[EN]Measuring semantic similarity and relatedness between textual items (words, sentences, paragraphs or even documents) is a very important research area in Natural Language Processing (NLP). In fact, it has many practical applications in other NLP tasks. For instance, Word Sense Disambiguation, Textual Entailment, Paraphrase detection, Machine Translation, Summarization and other related tasks such as Information Retrieval or Question Answering. In this masther thesis we study di erent approaches to compute the semantic similarity between textual items. In the framework of the european PATHS project1, we also evaluate a knowledge-base method on a dataset of cultural item descriptions. Additionaly, we describe the work carried out for the Semantic Textual Similarity (STS) shared task of SemEval-2012. This work has involved supporting the creation of datasets for similarity tasks, as well as the organization of the task itself.

Intranasal Delivery of Plasma and Platelet Growth Factors Using PRGF-Endoret System Enhances Neurogenesis in a Mouse Model of Alzheimer’s Disease

[EN] Neurodegeneration together with a reduction in neurogenesis are cardinal features of Alzheimer’s disease (AD) induced by a combination of toxic amyloid-β peptide (Aβ) and a loss of trophic factor support. Amelioration of these was assessed with diverse neurotrophins in experimental therapeutic approaches. The aim of this study was to investigate whether intranasal delivery of plasma rich in growth factors (PRGF-Endoret), an autologous pool of morphogens and proteins, could enhance hippocampal neurogenesis and reduce neurodegeneration in an amyloid precursor protein/presenilin-1 (APP/PS1) mouse model. Neurotrophic and neuroprotective actions were firstly evident in primary neuronal cultures, where cell proliferation and survival were augmented by Endoret treatment. Translation of these effects in vivo was assessed in wild type and APP/PS1 mice, where neurogenesis was evaluated using 5-bromodeoxyuridine (BdrU), doublecortin (DCX), and NeuN immunostaining 5 weeks after Endoret administration. The number of BrdU, DCX, and NeuN positive cell was increased after chronic treatment. The number of degenerating neurons, detected with fluoro Jade-B staining was reduced in Endoret-treated APP/PS1 mice at 5 week after intranasal administration. In conclusion, Endoret was able to activate neuronal progenitor cells, enhancing hippocampal neurogenesis, and to reduce Aβ-induced neurodegeneration in a mouse model of AD.

Analysis, optimization and development of an answer scoring system

The main contribution of this work is to analyze and describe the state of the art performance as regards answer scoring systems from the SemEval- 2013 task, as well as to continue with the development of an answer scoring system (EHU-ALM) developed in the University of the Basque Country. On the overall this master thesis focuses on finding any possible configuration that lets improve the results in the SemEval dataset by using attribute engineering techniques in order to find optimal feature subsets, along with trying different hierarchical configurations in order to analyze its performance against the traditional one versus all approach. Altogether, throughout the work we propose two alternative strategies: on the one hand, to improve the EHU-ALM system without changing the architecture, and, on the other hand, to improve the system adapting it to an hierarchical con- figuration. To build such new models we describe and use distinct attribute engineering, data preprocessing, and machine learning techniques.

System architecture and deployment scenarios for SESAME: Small cEllS coordinAtion for Multi-tenancy and Edge services

The surge of the Internet traffic with exabytes of data flowing over operators mobile networks has created the need to rethink the paradigms behind the design of the mobile network architecture. The inadequacy of the 4G UMTS Long term Evolution (LTE) and even of its advanced version LTE-A is evident, considering that the traffic will be extremely heterogeneous in the near future and ranging from 4K resolution TV to machine-type communications. To keep up with these changes, academia, industries and EU institutions have now engaged in the quest for new 5G technology. In this paper we present the innovative system design, concepts and visions developed by the 5G PPP H2020 project SESAME (Small cEllS coordinAtion for Multi-tenancy and Edge services). The innovation of SESAME is manifold: i) combine the key 5G small cells with cloud technology, ii) promote and develop the concept of Small Cellsas- a-Service (SCaaS), iii) bring computing and storage power at the mobile network edge through the development of nonx86 ARM technology enabled micro-servers, and iv) address a large number of scenarios and use cases applying mobile edge computing. Topics: