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.

Greenland turbot Reinhardtius hippoglossoides of the eastern Bering Sea and Aleutian Islands region

Greenland turbot (Reinhardtius hippoglossoides) is a commercially important flounder in both the North Atlantic and North Pacific Oceans. In the latter, its center of abundance is in the eastern Bering Sea and along the Aleutian Islands chain where its population is managed as a single stock. Harvest levels in this region of the North Pacific during the period 1970-81 were comparable with those in the northwest and northeast Atlantic, with annual average catches of 53,000 metric tons (t). However, the catch in 1984 dropped sharply to 23,100 t, in part because of reduced quotas arising from concern over continued poor recruitment and declining catch-per-unit-effort. Recruitment failure was manifested in 1) the sharp decline in the catch rate of young flsh in annual research trawl surveys on the continental shelf of the eastern Bering Sea and 2) an increasing proportion of older and larger fish in the commercial catch from the continental slope of both the eastern Bering Sea and Aleutian Islands. The cause ofthe decline in recruitment could not be clearly identifled. Greenland turbot of the Bering Sea-Aleutian Islands share certain distributional features with the North Atlantic form. There is an apparent bathymetric change in the size and age of fish, with younger animals occupying continental shelf depths and the older individuals residing at depths of the continental slope. At shallow depths the young are exposed to temperature fluctuations, whereas older animals along the slope are exposed to relatively stable temperatures. A hypothesis is proposed for describing the temporal and spatial paths by which young animals reach the mature or spawning portion of the population. (PDF file contains 38 pages.)

A histopathologic evaluation of gross lesions excised from commercially important North Altantic marine fishes

Histopathologic studies of lesions found in commercially important North Atlantic marine fishes are uncommon. As part of a comprehensive Northeast Fisheries Center program ("Ocean Pulse") to evaluate environmental and resource health on the U.S. Continental Shelf from Cape Hatteras to Nova Scotia, grossly visible lesions of the gills, integument, muscle, and viscera of primarily bottom-dwelling fishes were excised and examined using light microscopy. Several gadid and pleuronectid fishes accounted for most of the lesions observed. Most pathological examinations were incidental to samples taken for age and growth determination and evaluation of predator/prey relationships. Several gadids, with either gill, heart, or spleen lesions, were sampled more intensively. Gill lesions principally affected gadids and were caused by either microsporidans or an unidentified oocyte-like cell. The majority of gastrointestinal lesions consisted of encapsulated or encysted larval worms or microsporidan-induced cysts. Few heart lesions were found. Integumental lesioos included ulcers, lymphocystis, and trematode metacercariae. Liver lesions almost always consisted of encapsulated or encysted larval helminths. Necrotic granulomata were seen in muscle and microsporidan-induced granulomata in spleen. Although not numerous, histologically interesting lesions were noted in integument, heart, liver, spleen, and muscle of several fish species. Histologic study of tissues excised from a variety of demersal and pelagic fishes from the eastern North Atlantic (France, Germany, Spain) revealed assorted integumental, renal, hepatic, and splenic lesions. Small sample size and non-random sampling precluded obtaining a meaningful quantitative estimate of the prevalence of the observed lesions in the population at risk; however, a useful census has been made of the types of lesions present in commercially important marine fishes. (PDF file contains 20 pages.)

The feeding ecology of some zooplankters that are important prey items of larval fish

Diets of 76 species of fish larvae from most oceans of the world were inventoried on the basis of information in 40 published studies. Although certaln geographlc, size- and taxon-specific patterns were apparent, certain zooplankton taxa appeared in the diets of larvae of a variety of fish species in numerous localities. Included were six genera of calanoid copepods (Acartia, Calanus, Centropages, Paracalanus, Pseudocaianus, Temora), three genera of cyclopoid copepods (Corycaeus, Oilhona, Oncata), harpacticoid copepods, copepod nauplii, tintinoids, cladocerans of the genera Evadne and Podon, barnacle nauplii, gastropod larvae, pteropods of the genus Limacina, and appendicularians. Literature on feeding habits of these zooplankters reveals that most of the copepods are omnivorous, feeding upon both phytoplankton and other zooplankton. Some taxa, such as Calanus, Paracalanus, Pseudocalanus, and copepod nauplii appear to be primarily herbivorous, while others, such as Acartia, Centropages, Temora, and cyclopoids exhibit broad omnivory or carnivory. The noncopepod zooplankters are primarily filter-feeders upon pbytoplankton and/or bacterioplankton. Despite the importance of zooplankters in larval fish food webs, spectic knowledge of the feeding ecology of many taxa is poor. Further, much present knowledge comes only from laboratory investigations that may not accurately portray feeding habits of zooplankters in nature. Lack of knowledge of the feeding ecology of many abundant zooplankters, which are also important in larval fish food webs, precludes realistic understanding of pelagic ecosystem dynamics. (PDF file contains 34 pages.)

The decline in the commercially important species of fish and predominance of Clarias lazera in Lake Chad

About 86 species of fish have been recorded in the Lake Chad. Most of the species occurring in the lake are widespread including most of the commercially important species. Fish distribution in the lake was adversely affected by the 1972/73 and 1983/84 droughts. The commercially important species before the 1972/73 drought were; Lates, Labeo distichodus, Heterotis, Gymnarchus, Hydrocynus, Citharinus and Bagrus. Other species which had less commercial value at that time include, Clarias, Gnathanemus, Polypterus, Protopterus, Tilapia and Synodontis. Lates niloticus was the most predominant species of commercial importance comprising 50-60% of the total catches of fishermen between 1962 and 1973. Before the 1972/73 drought, occurrence of Clarias lazera was negligible and restricted to long line catches and had little commercial value. As a result of the drought of 1972/73 which resulted in near drying up to the northern sector of the lake, fish populations were confined to isolated pools and were completely scooped out. The remaining fish populations retreated to the southern basin where enough water always remained to hold the surviving representatives of the population. The effects of the drought resulted in occurrence of the little known C. lazera along with other hardy species like Tilapia and Protopterus to dominate the existing fish species composition. In 1976, C. lazera dominated the total catches of fishermen with 85.6%. An overview of the fish population in the lake, their relative abundance, changes in species composition, the effects of drought on the fauna of the lake based on the available data are discussed in this paper

Some aspects of the dynamics of populations important to the management of the commercial marine fisheries

(PDF contains 33 pages)

Prevalence of endoparasite in some commercially important freshwater fishes of Bida area, Niger State: a preliminary report

The result of this preliminary report highlighted prevalence of an endoparasite (nematode) in order of prevalence. Four non-scaly commercially important fishes, namely:- Clarias sp Hererobranchus bidorsalis. Bagrus sp and Synodontis sp.; and nine scaly fishes, namely; - Gymnarchus sp, Protopterus annectus. Tilapia sp, Lates niloticus. Heterotis niloticus, Mormyrops sp, Channa Obscurus, Labeo sp and Distochodus rostratus of freshwater fadama of the Bida Area, Niger State. The extent of the spread was evaluated. While the reactions of fishermen, middlemen (fishmongers) and fish-eaters in order to appreciate the impacts of the nematode infested fishes on the populace were recorded. The paper suggested areas for further studies towards evoking desirable management strategies for the study area as follows. Such suggested areas are:- Identification and confirmation of the nematode species; life cycle of the nematode species; influence of season on the prevalence, spread and pathology in the nematode infested fishes, etc

Analysis of a transcriptional network involving PU.1, Notch, and Gata3 in the lymphomyeloid lineage decision during early T-cell development

Hematopoiesis is a well-established system used to study developmental choices amongst cells with multiple lineage potentials, as well as the transcription factor network interactions that drive these developmental paths. Multipotent progenitors travel from the bone marrow to the thymus where T-cell development is initiated and these early T-cell precursors retain lineage plasticity even after initiating a T-cell program. The development of these early cells is driven by Notch signaling and the combinatorial expression of many transcription factors, several of which are also involved in the development of other cell lineages. The ETS family transcription factor PU.1 is involved in the development of progenitor, myeloid, and lymphoid cells, and can divert progenitor T-cells from the T-lineage to a myeloid lineage. This diversion of early T-cells by PU.1 can be blocked by Notch signaling. The PU.1 and Notch interaction creates a switch wherein PU.1 in the presence of Notch promotes T-cell identity and PU.1 in the absence of Notch signaling promotes a myeloid identity. Here we characterized an early T-cell cell line, Scid.adh.2c2, as a good model system for studying the myeloid vs. lymphoid developmental choice dependent on PU.1 and Notch signaling. We then used the Scid.adh.2c2 system to identify mechanisms mediating PU.1 and Notch signaling interactions during early T-cell development. We show that the mechanism by which Notch signaling is protecting pro-T cells is neither degradation nor modification of the PU.1 protein. Instead we give evidence that Notch signaling is blocking the PU.1-driven inhibition of a key set of T-regulatory genes including Myb, Tcf7, and Gata3. We show that the protection of Gata3 from PU.1-mediated inhibition, by Notch signaling and Myb, is important for retaining a T-lineage identity. We also discuss a PU.1-driven mechanism involving E-protein inhibition that leads to the inhibition of Notch target genes. This is mechanism may be used as a lockdown mechanism in pro-T-cells that have made the decision to divert to the myeloid pathway.

Photodissociation and reaction dynamics of chlorine-containing species important in stratospheric ozone chemistry

Chlorine oxide species have received considerable attention in recent years due to their central role in the balance of stratospheric ozone. Many questions pertaining to the behavior of such species still remain unanswered and plague the ability of researchers to develop accurate chemical models of the stratosphere. Presented in this thesis are three experiments that study various properties of some specific chlorine oxide species.

In the first chapter, the reaction between ClONO_2 and protonated water clusters is investigated to elucidate a possible reaction mechanism for the heterogeneous reaction of chlorine nitrate on ice. The ionic products were various forms of protonated nitric acid, NO_2 +(H_20)_m, m = 0, 1, 2. These products are analogous to products previously reported in the literature for the neutral reaction occurring on ice surfaces. Our results support the hypothesis that the heterogeneous reaction is acid-catalyzed.

In the second chapter, the photochemistry of ClONO_2 was investigated at two wavelengths, 193 and 248 nm, using the technique of photofragmentation translational spectroscopy. At both wavelengths, the predominant dissociation pathways were Cl + NO_3 and ClO + NO_2. Channel assignments were confirmed by momentum matching the counterfragments from each channel. A one-dimensional stratospheric model using the new 248 nm branching ratio determined how our results would affect the predicted Cl_x and NO_x partitioning in the stratosphere.

Chapter three explores the photodissociation dynamics of Cl_2O at 193, 248 and 308 nm. At 193 nm, we found evidence for the concerted reaction channel, Cl_2 + O. The ClO + Cl channel was also accessed, however, the majority of the ClO fragments were formed with sufficient internal energies for spontaneous secondary dissociation to occur. At 248 and 308 nm, we only observed only the ClO + Cl channel. . Some of the ClO formed at 248 nm was formed internally hot and spontaneously dissociated. Bimodal translational energy distributions of the ClO and Cl products indicate two pathways leading to the same product exist.

Appendix A, B and C discuss the details of data analysis techniques used in Chapters 1 and 2. The development of a molecular beam source of ClO dimer is presented in Appendix D.

Deciding to evacuate: Gathering the information to make this important decision

When hazardous storms threaten coastal communities, people need information to decide how to respond to this potential emergency. NOAA and NC Sea Grant are funding a two-year project (Risk Perceptions and Emergency Communication Effectiveness in Coastal Zones) to learn how residents, government officials, businesses and other organizations are informed and use information regarding hurricane and tropical storms. (PDF contains 4 pages)

An exposition on field identification of Clariid catfishes as an important tool in fish breeding and genetics

This paper conducts an exposition on the field identification of Clariid catfishes Heterobranchus, Clarias and their hybrid as an important tool in fish breeding and genetics. The paper explained the classification and aquacultural importance of Clariid catfishes. Parameters necessary for fish identification were highlighted. The identification of Heterobranchus, Clarias, their hybrid and sexual differences were also identified. The paper is of the position that the identification of Heterobranchus, Clarias and their hybrid is important in their genetic conservation and in achieving success in breeding and genetic studies

Seismic structure above and below the core-mantle boundary

Seismic structure above and below the core-mantle boundary (CMB) has been studied through use of travel time and waveform analyses of several different seismic wave groups. Anomalous systematic trends in observables document mantle heterogeneity on both large and small scales. Analog and digital data has been utilized, and in many cases the analog data has been optically scanned and digitized prior to analysis.

Differential travel times of S - SKS are shown to be an excellent diagnostic of anomalous lower mantle shear velocity (V s) structure. Wavepath geometries beneath the central Pacific exhibit large S- SKS travel time residuals (up to 10 sec), and are consistent with a large scale 0(1000 km) slower than average V_s region (≥3%). S - SKS times for paths traversing this region exhibit smaller scale patterns and trends 0(100 km) indicating V_s perturbations on many scale lengths. These times are compared to predictions of three tomographically derived aspherical models: MDLSH of Tanimoto [1990], model SH12_WM13 of Suet al. [1992], and model SH.10c.17 of Masters et al. [1992]. Qualitative agreement between the tomographic model predictions and observations is encouraging, varying from fair to good. However, inconsistencies are present and suggest anomalies in the lower mantle of scale length smaller than the present 2000+ km scale resolution of tomographic models. 2-D wave propagation experiments show the importance of inhomogeneous raypaths when considering lateral heterogeneities in the lowermost mantle.

A dataset of waveforms and differential travel times of S, ScS, and the arrival from the D" layer, Scd, provides evidence for a laterally varying V_s velocity discontinuity at the base of the mantle. Two different localized D" regions beneath the central Pacific have been investigated. Predictions from a model having a V_s discontinuity 180 km above the CMB agree well with observations for an eastern mid-Pacific CMB region. This thickness differs from V_s discontinuity thicknesses found in other regions, such as a localized region beneath the western Pacific, which average near 280 km. The "sharpness" of the V_s jump at the top of D", i.e., the depth range over which the V_s increase occurs, is not resolved by our data, and our data can in fact may be modeled equally well by a lower mantle with the increase in V_s at the top of D" occurring over a 100 krn depth range. It is difficult at present to correlate D" thicknesses from this study to overall lower mantle heterogeneity, due to uncertainties in the 3-D models, as well as poor coverage in maps of D" discontinuity thicknesses.

P-wave velocity structure (V_p) at the base of the mantle is explored using the seismic phases SKS and SPdKS. SPdKS is formed when SKS waves at distances around 107° are incident upon the CMB with a slowness that allows for coupling with diffracted P-waves at the base of the mantle. The P-wave diffraction occurs at both the SKS entrance and exit locations of the outer core. SP_dKS arrives slightly later in time than SKS, having a wave path through the mantle and core very close to SKS. The difference time between SKS and SP_dKS strongly depends on V_p at the base of the mantle near SK Score entrance and exit points. Observations from deep focus Fiji-Tonga events recorded by North American stations, and South American events recorded by European and Eurasian stations exhibit anomalously large SP_dKS - SKS difference times. SKS and the later arriving SP_dKS phase are separated by several seconds more than predictions made by 1-D reference models, such as the global average PREM [Dziewonski and Anderson, 1981] model. Models having a pronounced low-velocity zone (5%) in V_p in the bottom 50-100 km of the mantle predict the size of the observed SP_dK S-SKS anomalies. Raypath perturbations from lower mantle V_s structure may also be contributing to the observed anomalies.

Outer core structure is investigated using the family of SmKS (m=2,3,4) seismic waves. SmKS are waves that travel as S-waves in the mantle, P-waves in the core, and reflect (m-1) times on the underside of the CMB, and are well-suited for constraining outermost core V_p structure. This is due to closeness of the mantle paths and also the shallow depth range these waves travel in the outermost core. S3KS - S2KS and S4KS - S3KS differential travel times were measured using the cross-correlation method and compared to those from reflectivity synthetics created from core models of past studies. High quality recordings from a deep focus Java Sea event which sample the outer core beneath the northern Pacific, the Arctic, and northwestern North America (spanning 1/8th of the core's surface area), have SmKS wavepaths that traverse regions where lower mantle heterogeneity is pre- dieted small, and are well-modeled by the PREM core model, with possibly a small V_p decrease (1.5%) in the outermost 50 km of the core. Such a reduction implies chemical stratification in this 50 km zone, though this model feature is not uniquely resolved. Data having wave paths through areas of known D" heterogeneity (±2% and greater), such as the source-side of SmKS lower mantle paths from Fiji-Tonga to Eurasia and Africa, exhibit systematic SmKS differential time anomalies of up to several seconds. 2-D wave propagation experiments demonstrate how large scale lower mantle velocity perturbations can explain long wavelength behavior of such anomalous SmKS times. When improperly accounted for, lower mantle heterogeneity maps directly into core structure. Raypaths departing from homogeneity play an important role in producing SmKS anomalies. The existence of outermost core heterogeneity is difficult to resolve at present due to uncertainties in global lower mantle structure. Resolving a one-dimensional chemically stratified outermost core also remains difficult due to the same uncertainties. Restricting study to higher multiples of SmKS (m=2,3,4) can help reduce the affect of mantle heterogeneity due to the closeness of the mantle legs of the wavepaths. SmKS waves are ideal in providing additional information on the details of lower mantle heterogeneity.

Transceiver designs and analysis for LTI, LTV and broadcast channels - new matrix decompositions and majorization theory

Signal processing techniques play important roles in the design of digital communication systems. These include information manipulation, transmitter signal processing, channel estimation, channel equalization and receiver signal processing. By interacting with communication theory and system implementing technologies, signal processing specialists develop efficient schemes for various communication problems by wisely exploiting various mathematical tools such as analysis, probability theory, matrix theory, optimization theory, and many others. In recent years, researchers realized that multiple-input multiple-output (MIMO) channel models are applicable to a wide range of different physical communications channels. Using the elegant matrix-vector notations, many MIMO transceiver (including the precoder and equalizer) design problems can be solved by matrix and optimization theory. Furthermore, the researchers showed that the majorization theory and matrix decompositions, such as singular value decomposition (SVD), geometric mean decomposition (GMD) and generalized triangular decomposition (GTD), provide unified frameworks for solving many of the point-to-point MIMO transceiver design problems.

In this thesis, we consider the transceiver design problems for linear time invariant (LTI) flat MIMO channels, linear time-varying narrowband MIMO channels, flat MIMO broadcast channels, and doubly selective scalar channels. Additionally, the channel estimation problem is also considered. The main contributions of this dissertation are the development of new matrix decompositions, and the uses of the matrix decompositions and majorization theory toward the practical transmit-receive scheme designs for transceiver optimization problems. Elegant solutions are obtained, novel transceiver structures are developed, ingenious algorithms are proposed, and performance analyses are derived.

The first part of the thesis focuses on transceiver design with LTI flat MIMO channels. We propose a novel matrix decomposition which decomposes a complex matrix as a product of several sets of semi-unitary matrices and upper triangular matrices in an iterative manner. The complexity of the new decomposition, generalized geometric mean decomposition (GGMD), is always less than or equal to that of geometric mean decomposition (GMD). The optimal GGMD parameters which yield the minimal complexity are derived. Based on the channel state information (CSI) at both the transmitter (CSIT) and receiver (CSIR), GGMD is used to design a butterfly structured decision feedback equalizer (DFE) MIMO transceiver which achieves the minimum average mean square error (MSE) under the total transmit power constraint. A novel iterative receiving detection algorithm for the specific receiver is also proposed. For the application to cyclic prefix (CP) systems in which the SVD of the equivalent channel matrix can be easily computed, the proposed GGMD transceiver has K/log_2(K) times complexity advantage over the GMD transceiver, where K is the number of data symbols per data block and is a power of 2. The performance analysis shows that the GGMD DFE transceiver can convert a MIMO channel into a set of parallel subchannels with the same bias and signal to interference plus noise ratios (SINRs). Hence, the average bit rate error (BER) is automatically minimized without the need for bit allocation. Moreover, the proposed transceiver can achieve the channel capacity simply by applying independent scalar Gaussian codes of the same rate at subchannels.

In the second part of the thesis, we focus on MIMO transceiver design for slowly time-varying MIMO channels with zero-forcing or MMSE criterion. Even though the GGMD/GMD DFE transceivers work for slowly time-varying MIMO channels by exploiting the instantaneous CSI at both ends, their performance is by no means optimal since the temporal diversity of the time-varying channels is not exploited. Based on the GTD, we develop space-time GTD (ST-GTD) for the decomposition of linear time-varying flat MIMO channels. Under the assumption that CSIT, CSIR and channel prediction are available, by using the proposed ST-GTD, we develop space-time geometric mean decomposition (ST-GMD) DFE transceivers under the zero-forcing or MMSE criterion. Under perfect channel prediction, the new system minimizes both the average MSE at the detector in each space-time (ST) block (which consists of several coherence blocks), and the average per ST-block BER in the moderate high SNR region. Moreover, the ST-GMD DFE transceiver designed under an MMSE criterion maximizes Gaussian mutual information over the equivalent channel seen by each ST-block. In general, the newly proposed transceivers perform better than the GGMD-based systems since the super-imposed temporal precoder is able to exploit the temporal diversity of time-varying channels. For practical applications, a novel ST-GTD based system which does not require channel prediction but shares the same asymptotic BER performance with the ST-GMD DFE transceiver is also proposed.

The third part of the thesis considers two quality of service (QoS) transceiver design problems for flat MIMO broadcast channels. The first one is the power minimization problem (min-power) with a total bitrate constraint and per-stream BER constraints. The second problem is the rate maximization problem (max-rate) with a total transmit power constraint and per-stream BER constraints. Exploiting a particular class of joint triangularization (JT), we are able to jointly optimize the bit allocation and the broadcast DFE transceiver for the min-power and max-rate problems. The resulting optimal designs are called the minimum power JT broadcast DFE transceiver (MPJT) and maximum rate JT broadcast DFE transceiver (MRJT), respectively. In addition to the optimal designs, two suboptimal designs based on QR decomposition are proposed. They are realizable for arbitrary number of users.

Finally, we investigate the design of a discrete Fourier transform (DFT) modulated filterbank transceiver (DFT-FBT) with LTV scalar channels. For both cases with known LTV channels and unknown wide sense stationary uncorrelated scattering (WSSUS) statistical channels, we show how to optimize the transmitting and receiving prototypes of a DFT-FBT such that the SINR at the receiver is maximized. Also, a novel pilot-aided subspace channel estimation algorithm is proposed for the orthogonal frequency division multiplexing (OFDM) systems with quasi-stationary multi-path Rayleigh fading channels. Using the concept of a difference co-array, the new technique can construct M^2 co-pilots from M physical pilot tones with alternating pilot placement. Subspace methods, such as MUSIC and ESPRIT, can be used to estimate the multipath delays and the number of identifiable paths is up to O(M^2), theoretically. With the delay information, a MMSE estimator for frequency response is derived. It is shown through simulations that the proposed method outperforms the conventional subspace channel estimator when the number of multipaths is greater than or equal to the number of physical pilots minus one.