Online Comment Text Analysis with Improved Feature Weight

Word vector is an important tool for natural language processing (NLP) tasks such as text classification. However, existing static language models such as Word2vec cannot solve the polysemy problem, leading to a decline in text classification performance. To solve this problem, this paper proposes a method for making Chinese word vector dynamic (MCWVD). The part of speech (POS) is used to solve the ambiguity problem caused by different POS. The POS structure graph is constructed and the syntactic structure information of POS features is extracted by GCN (Graph Convolutional Network). POS vector and word vector are concatenated into PW (POS-Word) vector. Parametric matrix is added to improve the fusion effect of POS and word features. Multilayer attention is used to distinguish the importance of different features and further update the vector expression of word vectors about the current context. Experiments on Chinese datasets THUCNews and SogouNews show that MCWVD effectively improves the accuracy of text classification and achieves better performance than CoVe (Context Vectors) and ELMo (Embeddings from Language Models). MCWVD also achieves similar performance to BERT and GPT-1 (Generative Pre-Training), but with a much lower computational cost and only 4% of BERT parameters.

Many applications require semantic understanding of short texts, and inferring discriminative and coherent latent topics is a critical and fundamental task in these applications. Conventional topic models largely rely on word co-occurrences to derive topics from a collection of documents. However, due to the length of each document, short texts are much more sparse in terms of word co-occurrences. Recent studies show that the Dirichlet Multinomial Mixture (DMM) model is effective for topic inference over short texts by assuming that each piece of short text is generated by a single topic. However, DMM has two main limitations. First, even though it seems reasonable to assume that each short text has only one topic because of its shortness, the definition of “shortness” is subjective and the length of the short texts is dataset dependent. That is, the single-topic assumption may be too strong for some datasets. To address this limitation, we propose to model the topic number as a Poisson distribution, allowing each short text to be associated with a small number of topics (e.g., one to three topics). This model is named PDMM. Second, DMM (and also PDMM) does not have access to background knowledge (e.g., semantic relations between words) when modeling short texts. When a human being interprets a piece of short text, the understanding is not solely based on its content words, but also their semantic relations. Recent advances in word embeddings offer effective learning of word semantic relations from a large corpus. Such auxiliary word embeddings enable us to address the second limitation. To this end, we propose to promote the semantically related words under the same topic during the sampling process, by using the generalized Pólya urn (GPU) model. Through the GPU model, background knowledge about word semantic relations learned from millions of external documents can be easily exploited to improve topic modeling for short texts. By directly extending the PDMM model with the GPU model, we propose two more effective topic models for short texts, named GPU-DMM and GPU-PDMM. Through extensive experiments on two real-world short text collections in two languages, we demonstrate that PDMM achieves better topic representations than state-of-the-art models, measured by topic coherence. The learned topic representation leads to better accuracy in a text classification task, as an indirect evaluation. Both GPU-DMM and GPU-PDMM further improve topic coherence and text classification accuracy. GPU-PDMM outperforms GPU-DMM at the price of higher computational costs.

The problem of this research is the learning plan, the implementation of learning, and the learning result of analyzing short story text using the Hypno-teaching method on the students of class XI SMAN 15 Bandung. The method of this research is descriptive. The research technique is an observation and test sheet. The population of this study is all the results of the ability to analyze the intrinsic element of short stories of students from XI MIPA 1 to XI MIPA 5 in SMAN 15 Kota Bandung, amounting to 215 students, and the sample is amounting to 31 students of class XI MIPA 5. The result of the research shows that: 1) learning planning to analyze short story text by using the Hypno-teaching method prepared by the teacher is appropriate, 2) implementation of learning done by teacher in accordance with RPP, 3) learning result analyze short story text using the Hypno-teaching method indicated from preliminary tests with an average score of 66.80 or 66.80%, and the final test score averaged 78.23 or 78.23%. The increase is 11.42%. In conclusion, effectiveness the Hypno-teaching method is used as a learning method in analyzing short stories.

PERCH SELECTION IN MALE DRAGONFLIES (ODONATA, ANISOPTERA) RELATED TO COMPETITIVE ABILITY AND SPECIES COMPOSITION Jessica L. Beard Old Dominion University, 2015 Director: Dr. Deborah Waller Males of many species of dragonflies (Odonata, Anisoptera) establish territories in aquatic habitats where they compete with other males for access to food and females. Territorial males typically perch on emergent vegetation and chase rival males who intrude into their territories. This dissertation research examined the role of male size in perch height selection, position on the perch, and competitive ability. Four hypotheses were tested: 1) Dragonfly species would vary by size and that territorial species would show sexual size dimorphism (SSD), 2) Perch height selection would be related to dragonfly size, 3) Position on the perch would be related to male size, with larger males selecting perch tops and smaller inferior competitors choosing the sides of perches, and 4) Intraspecific competition would be more important than interspecific competition. Research was conducted at four lakes in southeastern Virginia from 2011-2014. For size measurements, male and female dragonflies were captured and measured for total body length, abdomen length, cerci length, forewing length and width, hindwing length and width and fresh mass. For perching experiments, alternating short (30cm above waterline) and tall (90cm above waterline) bamboo perches were placed in two rows, 0.5m and 2.0m from the shore. Any dragonflies that alighted on perches were recorded for species, gender, perch position and length of occupancy. Any interactions with conspecific or heterospecific dragonflies were recorded. Results showed that dragonfly males varied significantly among species in all parameters measured, and SSD was found for some parameters for some of the species. In particular, females of several species had greater forewing and hindwing widths than males, perhaps related to selection for energy conservation in females. There was no association between dragonfly size and perch height selection. Four species frequently perched on the sides rather than the tops of

A Heterogeneous Directed Graph Attention Network for inductive text classification using multilevel semantic embeddings

This research background because there are still many students who are not interested in writing short stories and the author has an idea to create the ability and willingness so that students interested in writing short stories text. The purpose of this research is to know the effectiveness of problem-based learning method. The method used in this study is a quasi-experimental method with the type of non-equivalent control group design, Based on the observer's observation that 100% of teachers are able to implement the process of writing short story text using problem-based learning method well. Based on the assessment by observer that 93.3% of students are able to follow the learning process of reading exposition text using problem-based learning method well. the results of the initial test and final tests in both classes obtained the average value of the final test of the experimental class of 77, the control class to obtain the average final test score of 68. Manwhitney test hypothesis test obtained 0.000 this means Ho is rejected and Ha accepted meaning there is a significant difference in the ability to write short story text using problem-based learning method. Student response showed positive response, evidenced by the number of positive statements of students 62% and 38% negative statements.) Population in this study is all students of class XI SMK Citra Pembaharuan with two class samples of 60 students.

With the rapid development of information technology and the Internet, text information on various media platforms, such as blogs, microblogs, and news, has exploded. In order to process these text data more efficiently, text classification technology has gradually become a research focus. In view of the common problems of incomplete structure and sparse features in short texts, this paper designs a short text classification model with multi-level semantic feature fusion to effectively extract key word features and sentence features from short texts. The model first uses BERT for preliminary semantic feature extraction, and then combines the first-layer Bi-GRU and the word-level self-attention mechanism to enhance the weight of important semantic information in the text. Next, the second-layer Bi-GRU and the sentence-level self-attention mechanism are used to achieve feature fusion at the word vector and sentence vector levels. The maximum pooling strategy of CNN is used to screen the best features and predict the category probability. In addition, the loss function is adjusted, a threshold is introduced to screen and continuously train samples that are difficult to classify, and an adaptive adjustment factor is added in the back propagation to adaptively attenuate the learning rate and accelerate the model fitting process. In the experiment of 20 Newsgroups dataset, compared with the baseline model, the classification accuracy of this model increased by 3.51% to 88.74%, which effectively proved the effectiveness of this model in improving the classification accuracy of Chinese short texts.

Fiber-reinforced polymer (FRP) composites offer several advantages in relation to traditional materials, such as high specific strength, good corrosion resistance, low thermal conductivity and rapid component installation. Despite the great potential of these materials, two major disadvantages limit their acceptance in civil engineering applications: their lack of inherent ductility and the fact that their fibrous and anisotropic character makes the joining of structural components difficult. The purposes of this research are to develop ductile structures using brittle all-FRP materials and to provide a corresponding design method assuring robustness and structural safety. The proposed concept for all-FRP structures incorporates system ductility through the use of ductile, adhesively-bonded joints and redundant (statically indeterminate) structural systems. The concept envisages adhesives with an initially elastic behavior sufficiently stiff to meet short- and long-term serviceability requirements. When serviceability and ultimate loads are exceeded, however, adhesive behavior should change and become plastic, or at least highly non-linear inelastic, with much lower stiffness. The ductile connections effect a favorable redistribution of internal and external forces in the redundant system in the same way as plastic hinges in statically indeterminate systems with ductile materials. In case of joint failure, the redundant system allows the creation of alternative load paths and redistribution of section forces preventing structural collapse. Numerical and experimental investigations of adhesively-bonded double-lap joints demonstrate that using an appropriate bilinear adhesive is more advantageous than using a traditional structural and stiff epoxy adhesive. These investigations show that ductile adhesives generate a linear axial strain distribution and consequently a constant shear stress distribution along the overlap. Numerical and analytical modelings are validated with experimental results. The numerical modeling allows evaluation of the effect of ductile adhesive behavior on joint behavior and establishment of recommendations for the selection of adhesive mechanical properties, according to serviceability and failure limit state requirements. Based on numerical modeling results, joints can also be classified according to stiffness (stiff or flexible). An analytical model, based on numerical and experimental results, is developed for predicting joint stiffness when using ductile adhesives, assuming that the adhesive deforms only and uniformly in shear and that adherends deform only in tension. The design method is developed in accordance with the new structural concept and Eurocode design philosophy. It is based on both numerical and analytical model results for adhesively-bonded double-lap joints. A case study illustrating the adhesively-bonded joint design and appropriate adhesive selection is carried out on continuous beams with adhesively-bonded joints at mid-support. Corresponding bending experiments on beams with ductile, adhesively-bonded connections demonstrate the new structural concept's feasibility and validate the proposed design method. An energy factor defined as the ultimate strength to serviceability limit state energy ratio is proposed to quantify and compare the robustness and safety of structures built from both ductile and brittle materials.

Drivers of Departure and Prospecting in Dispersing Juvenile Red Kites (Milvus milvus)

This paper presents an ethologically inspired actionselection mechanism, part of a larger architecture for selfanimated artificial animals (agents) that communicate emotions to each other, influencing each other’s behaviour. The problem of action selection is that of choosing at each moment in time the most appropriate action out of a repertoire of possible actions. Action selection algorithms have been proposed by both ethologists and computer scientists. Models suggested by ethologists are usually at a conceptual level, while those of computer scientists (with some exceptions [3] ) generally do not take into account classical ethology theories. According several ethologists [2], a hierarchical structure represents an essential organising principle of complex behaviours. We have chosen to add emotion to a hierarchical action selection mechanism so that behaviour shows persistence (emotion acting as a short term memory [2]) and to avoid dithering between competing behaviours, namely the herding group behaviour, and the individual behaviour of grazing.

A short text topic modeling method based on integrating Gaussian and Logistic coding networks with pre-trained word embeddings

As one of the core tasks of natural language processing technology, text classification methods general-ly face the problems of insufficient global semantic capture and limited feature focusing ability when processing long texts or complex semantics. To address this issue, a deep learning model that integrates improved convolutional neural networks, unbalanced bidirectional long short-term memory networks, and multi-head attention mechanisms is proposed. Utilizing an improved bidirectional long short-term memory network to capture global semantic information, while dynamically focusing on key features through a multi head attention mechanism to enhance the model's adaptability to classification tasks. The performance of the model is validated through experiments on AG News (short text) and IMDb (long text) datasets. The results show that in short text classification, the proposed method has an accu-racy rate of 96% and a classification error rate of only 1.46%. In the task of long text classification, the method proposed in the study has a product under the curve of 0.98. In adversarial attack testing, the accuracy rates of adversarial samples generated by different methods are 92.85% and 90.63%, respec-tively, with the lowest robustness degradation rates of 3.72% and 5.49%, respectively. In cross domain generalization testing, it shows the least classification errors and superior cross domain adaptability. These results validate the high performance, robustness, and wide applicability of the method. The re-search indicates that this approach can validly improve the performance of text classification and pro-vide new solutions for natural language processing related tasks in long text and multi-category scenar-ios.

As a social product in the Internet age, Microblog has the characteristics of concise language and large amount of information. Mining the information in the Microblog text is of great significance to topic detection and personalized recommendation systems. Compared with traditional text classification, short text has the problems of dimensionality and sparseness. The article first uses the neural network's Word2Vec to get the word vector, and then introduces the category attribute into the word weight. The short text vector is obtained by the weighted summation method, and finally the text data is trained using the SVM classifier. Through the comparison of experimental results, the effectiveness of the method is verified.

Word embeddings act as an important component of deep models for providing input features in downstream language tasks, such as sequence labelling and text classification. In the last decade, a substantial number of word embedding methods have been proposed for this purpose, mainly falling into the categories of classic and context-based word embeddings. In this paper, we conduct controlled experiments to systematically examine both classic and contextualised word embeddings for the purposes of text classification. To encode a sequence from word representations, we apply two encoders, namely CNN and BiLSTM, in the downstream network architecture. To study the impact of word embeddings on different datasets, we select four benchmarking classification datasets with varying average sample length, comprising both single-label and multi-label classification tasks. The evaluation results with confidence intervals indicate that CNN as the downstream encoder outperforms BiLSTM in most situations, especially for document context-insensitive datasets. This study recommends choosing CNN over BiLSTM for document classification datasets where the context in sequence is not as indicative of class membership as sentence datasets. For word embeddings, concatenation of multiple classic embeddings or increasing their size does not lead to a statistically significant difference in performance despite a slight improvement in some cases. For context-based embeddings, we studied both ELMo and BERT. The results show that BERT overall outperforms ELMo, especially for long document datasets. Compared with classic embeddings, both achieve an improved performance for short datasets while the improvement is not observed in longer datasets.

A waste rock characterization effort was designed and implemented to identify favorable waste rock borrow material for use as rock armor on a tailing impoundment side-slope reclamation project. The following criteria were applied: 1) sufficient volumes of material; 2) particle size distributions with greater than 50% of the material larger than 12.5-mm particle diameter; 3) acid-base accounting showing positive net neutralization potential; and 4) sufficient nutrient levels and minimal potentially phytotoxic elements to facilitate revegetation. In addition, a field-scale erosion test designed to simulate a 100-year, 24-hour storm event was conducted to determine the efficacy of the selected rock material and of selected ripping treatments in controlling erosion. Two generally non-acid generating and erosion resistant mine-waste rock types were found: argillite and arkose material. Within the potential source areas, evidence of oxidation (yellowish-red staining) was a good predictor of net acid -generating material and used as exclusion criteria. Argillite was used as rock armor on short slopes and the coarser arkose material for slope-lengths greater than 90-meters.