Extracting Topics with Simultaneous Word Co-occurrence and Semantic Correlation Graphs: Neural Topic Modeling for Short Texts

Short text nowadays has become a more fashionable form of text data, e.g., Twitter posts, news titles, and product reviews. Extracting semantic topics from short texts plays a significant role in a wide spectrum of NLP applications, and neural topic modeling is now a major tool to achieve it. Motivated by learning more coherent and semantic topics, in this paper we develop a novel neural topic model named Dual Word Graph Topic Model (DWGTM), which extracts topics from simultaneous word co-occurrence and semantic correlation graphs. To be specific, we learn word features from the global word co-occurrence graph, so as to ingest rich word co-occurrence information; we then generate text features with word features, and feed them into an encoder network to get topic proportions per-text; finally, we reconstruct texts and word co-occurrence graph with topical distributions and word features, respectively. Besides, to capture semantics of words, we also apply word features to reconstruct a word semantic correlation graph computed by pre-trained word embeddings. Upon those ideas, we formulate DWGTM in an auto-encoding paradigm and efficiently train it with the spirit of neural variational inference. Empirical results validate that DWGTM can generate more semantically coherent topics than baseline topic models.

Modeling Diagnostic Label Correlation for Automatic ICD Coding

Given the clinical notes written in electronic health records (EHRs), it is challenging to predict the diagnostic codes which is formulated as a multi-label classification task. The large set of labels, the hierarchical dependency, and the imbalanced data make this prediction task extremely hard. Most existing work built a binary prediction for each label independently, ignoring the dependencies between labels. To address this problem, we propose a two-stage framework to improve automatic ICD coding by capturing the label correlation. Specifically, we train a label set distribution estimator to rescore the probability of each label set candidate generated by a base predictor. This paper is the first attempt at learning the label set distribution as a reranking module for ICD coding. In the experiments, our proposed framework is able to improve upon best-performing predictors for medical code prediction on the benchmark MIMIC datasets.

The impact of industrial demographic variables on economic sectors in Syria

This research aims at studying the relationship between the industrial demographic variables as it is one of the most important Syrian economic sectors and the most laboremployment, and the inputs of this sector in particular and the inputs of the rest of the economic sectors. A number of results were found, most notably the existence of a statistically significant relationship between the inputs of the industrial sector and the total number of workers in the industrial sector, and the relationship between these inputs and the qualitative, sectoral and educational structure of the industrial sector.

Constructing a geological model and its use in deepening the oil understanding of the formations of Ruttba and Malussa (F) in Sabban Field - low Euphrates

Due to the wide and rapid processing provided by the petrel program, we have relied on the construction of a three-dimensional statistical model of the studied area (Sabban Field - Ruttba and Malussa Formations) using data for well-studied wells (geochemistry, geophysical measurements and interpreted seismic data) To complete the reservoir study in a subsequent study.

Genetic and phenotypic Analysis of Some Productivity and Qualitative Traits of Some Syrian cotton Genotypes in Hama Governorate

Nine Syrian cotton genotypes i.e. Aleppo 33/1, Aleppo 118, Aleppo 90, Aleppo 40, Aleppo 124 Rakka5, Deir Al-Zour 22, Line 106 and Rusafa, were used for statistical and genetic analysis for productivity and some chemical indicators of seeds: Lint percentage%, cotton weight per plant/g, percent of oil in seeds and pulp, percent of protein in seeds and pulp, percent of humidity in seeds and pulp, to explore the potentiality of the genotypes in the studied region, and to establish a program for the production of cotton and seeds, also to determine the selective indecies to be used to improve cotton productivity and seed components, using randomized complete block design with three replications. The experiment was conducted in Salhab village, Al-Ghab region, Hama governorate, Syria, during 2015 season.

Path Coefficient Analysis and Relative Importance of Some Fodder Yield Components in Pearl millet [Pennisetum glaucum (L.) R. Br.]

This study was carried out at Kharabu Station, General Commission for Scientific Agricultural Research (GCSAR), during the growing seasons 2011 and 2012, to estimate phenotypic correlation, and path coefficient of some morphological and green fodder yield components (days to flowering, plant height (cm), number of leaves per plant, number of tillers per plant, and green fodder yield (ton/ha)). Fifteen pearl millet crosses resulted from half diallel mating system between six inbred lines. The experiment was arranged in a randomized complete block design (RCBD), with three replications.

Inheritance of grain yield and its components in line by tester method of maize (Zea mays L.)

This study was carried out at the Scientific Agriculture Research Center, Al-Ghab, Syria, during 2013 –2014 growing seasons to estimate heterosis, combining ability, phenotypic correlation and path analysis for plant and ear height, ear length, ear diameter, number of rows per ear, number of kernels per row, 100 kernel weight and grain yield per plant for eighteen hybrids produced by the line × tester method, the major findings were: inbred lines, testers, hybrids and combining ability mean squares were significant for all traits, indicating that additive and non-additive gene actions were the important in inheritance of all traits. The ratios of σ2 GCA/σ2 SCA showed that non-additive gene action was more important in controlling all traits except of number of kernel per row. Heterosis percentage for all traits were significant compared with the check variety except of ear height trait. GCA effects showed that the lines P1 and P7 were good combiners for grain yield per plant, also, SCA effects showed that P2×P8, P5×P8, P4×P9 and P1×P7 crosses were the best F1 combiners for grain yield per plant. Results of phenotypic correlation and path analysis values showed that ear length, ear diameter and number of kernel per row were positively and significantly associated with grain yield per plant, also, these traits can be considered as selection criteria may lead to the improvement of grain yield in maize.

Evaluation of machines maintenance depending on key performance indicators

This research includes evaluating the work of maintenance of themachines through their own key performance indicators (KPI), and approved byindustrial and service companies, We mention reliability, readiness ,average time between failures, average Repair time ,average time of periodic maintenance of their production and serviceequipment. After the data processing of emergency failures and periodic maintenanceof machines , we extracted values for maintenance performanceindicators, we measured their performance and correlation to evaluateits compatibility with global indicators. The research aim to assess the current situation through the use ofquantitative methods pattern which is used to access the values of theprevious mentioned performance indicators and their interdependence tobe used in support of maintenance decisionsin the container terminal later. We noticed a strong inverse association between periodic maintenance TPMtime and breakdown TBD, and reached a value rERS = -0.99, and thereforewe must focus on increasing periodic maintenance to reduce breakdowntime, and replace parts that cause recurringproblems which caused about intensive breakdowns, thus reducing periodicmaintenance time. Also we were able to classify the quality of maintenance and machinesthrough the readiness and reliability indicators and the mean timebetween failures.

Estimation of Correlation and Path coefficient for hybrids of maize(Zea mays) under conditions industrial infection by large corn stem borer Sesamia cretica

It was evaluated the susceptibility of 6 lines of maize, and fifteen maize crosses, against by large corn stem borer, Sesamia cretica Led under conditions Industrial infection. A study of the correlation coefficient showed linear relationship of yields by 100- Kernel weight.

Suggestion of Cost Index For School Building Projects in Lattakia

Cost indices are considered as important tools which help both the owner and the contractor to identify the primary evaluation of project depending on the cost of previous and similar project in easy and quick form, and the most important using ways by researchers in process modernization previous cost in shortest time and least effort. Methodology was adopted to help in designing cost index taking into account weights of chosen groups instead of weights of items, so It has been suggested three cost indices for school building in Lattakia by studying a sample consists of 32 schools were constructed between 2001-2012, and then The best cost index between them which explains the biggest portion of square meter cost variations due to escalation was selected, also it has been suggested a model that reflect the relationship between cost of square meter and developed cost index for corresponding year through statically program spss.