Domain adaption for word segmentation and POS tagging is a challenging problem for Chinese lexical processing. Self-training is one promising solution for it, which struggles to construct a set of high-quality pseudo training instances for the target
domain. Previous work usually assumes a universal source-to-target adaption to collect such pseudo corpus, ignoring the different gaps from the target sentences to the source domain. In this work, we start from joint word segmentation and POS tagging, presenting a fine-grained domain adaption method to model the gaps accurately. We measure the gaps by one simple and intuitive metric, and adopt it to develop a pseudo target domain corpus based on fine-grained subdomains incrementally. A novel domain-mixed representation learning model is proposed accordingly to encode the multiple subdomains effectively. The whole process is performed progressively for both corpus construction and model training. Experimental results on a benchmark dataset show that our method can gain significant improvements over a vary of baselines. Extensive analyses are performed to show the advantages of our final domain adaption model as well.
Intelligent agents that are confronted with novel concepts in situated environments will need to ask their human teammates questions to learn about the physical world. To better understand this problem, we need data about asking questions in situated
task-based interactions. To this end, we present the Human-Robot Dialogue Learning (HuRDL) Corpus - a novel dialogue corpus collected in an online interactive virtual environment in which human participants play the role of a robot performing a collaborative tool-organization task. We describe the corpus data and a corresponding annotation scheme to offer insight into the form and content of questions that humans ask to facilitate learning in a situated environment. We provide the corpus as an empirically-grounded resource for improving question generation in situated intelligent agents.
The outstanding performance of transformer-based language models on a great variety of NLP and NLU tasks has stimulated interest in exploration of their inner workings. Recent research has been primarily focused on higher-level and complex linguistic
phenomena such as syntax, semantics, world knowledge and common-sense. The majority of the studies is anglocentric, and little remains known regarding other languages, specifically their morphosyntactic properties. To this end, our work presents Morph Call, a suite of 46 probing tasks for four Indo-European languages of different morphology: Russian, French, English and German. We propose a new type of probing tasks based on detection of guided sentence perturbations. We use a combination of neuron-, layer- and representation-level introspection techniques to analyze the morphosyntactic content of four multilingual transformers, including their understudied distilled versions. Besides, we examine how fine-tuning on POS-tagging task affects the probing performance.
Transformer has achieved great success in the NLP field by composing various advanced models like BERT and GPT. However, Transformer and its existing variants may not be optimal in capturing token distances because the position or distance embeddings
used by these methods usually cannot keep the precise information of real distances, which may not be beneficial for modeling the orders and relations of contexts. In this paper, we propose DA-Transformer, which is a distance-aware Transformer that can exploit the real distance. We propose to incorporate the real distances between tokens to re-scale the raw self-attention weights, which are computed by the relevance between attention query and key. Concretely, in different self-attention heads the relative distance between each pair of tokens is weighted by different learnable parameters, which control the different preferences on long- or short-term information of these heads. Since the raw weighted real distances may not be optimal for adjusting self-attention weights, we propose a learnable sigmoid function to map them into re-scaled coefficients that have proper ranges. We first clip the raw self-attention weights via the ReLU function to keep non-negativity and introduce sparsity, and then multiply them with the re-scaled coefficients to encode real distance information into self-attention. Extensive experiments on five benchmark datasets show that DA-Transformer can effectively improve the performance of many tasks and outperform the vanilla Transformer and its several variants.
Sentiment analysis has attracted increasing attention in e-commerce. The sentiment polarities underlying user reviews are of great value for business intelligence. Aspect category sentiment analysis (ACSA) and review rating prediction (RP) are two es
sential tasks to detect the fine-to-coarse sentiment polarities. ACSA and RP are highly correlated and usually employed jointly in real-world e-commerce scenarios. While most public datasets are constructed for ACSA and RP separately, which may limit the further exploitation of both tasks. To address the problem and advance related researches, we present a large-scale Chinese restaurant review dataset ASAP including 46, 730 genuine reviews from a leading online-to-offline (O2O) e-commerce platform in China. Besides a 5-star scale rating, each review is manually annotated according to its sentiment polarities towards 18 pre-defined aspect categories. We hope the release of the dataset could shed some light on the field of sentiment analysis. Moreover, we propose an intuitive yet effective joint model for ACSA and RP. Experimental results demonstrate that the joint model outperforms state-of-the-art baselines on both tasks.
Deciding whether a semantically ambiguous word is homonymous or polysemous is equivalent to establishing whether it has any pair of senses that are semantically unrelated. We present novel methods for this task that leverage information from multilin
gual lexical resources. We formally prove the theoretical properties that provide the foundation for our methods. In particular, we show how the One Homonym Per Translation hypothesis of Hauer and Kondrak (2020a) follows from the synset properties formulated by Hauer and Kondrak (2020b). Experimental evaluation shows that our approach sets a new state of the art for homonymy detection.
IOT sensors use the publish/subscribe model for communication to benefit from its
decoupled nature with respect to space, time, and synchronization. Because of the
heterogeneity of communicating parties, semantic decoupling is added as a fourth
di
mension. The added semantic decoupling complicates the matching process and reduces
its efficiency. The proposed algorithm clusters subscriptions and events according to topic
and performs the matching process within these clusters, which increases the throughput
by reducing the matching time . Moreover, the accuracy of matching is improved when
subscriptions must be fully approximated . This work shows the benefit of clustering, as
well as the improvement in the matching accuracy and efficiency achieved using this
approach.
Introduction: Liver surgery is a relatively recent and relatively major complex surgery, due
to the anatomical and histological specialty of the liver
Objective of the research: To evaluate the results of the cases of liver metastases from
colorec
tal carcinoma from November 2010 until the end of April 2017.The study included
11 patients with 63,4% of females and average age of 55,4 years.
The primary tumor was located in the rectum at 36.4%. in 4 cases (36.4%) was the
livermetastases as a synchron with the main tumor while the metachron metastases was in
63.6% of cases. The single metastases was found at 72.7% while the multiple metastases
accounted for 27.3%. The transfusion diameter was less than 5 cm in 5 cases and 45.4%
and it was larger or equal to 5 cm in 54.6% of all the cases. The liver resection for
Hepatocellular lesions were performed concurrently with primary tumor resection in the
colon in only one case and the metastases was a single and 9.1% of the cases. The most
common surgical procedure was segment resection in 45.4% of all the cases. The left or
right hemi hepatectomy was in two cases, 8.7% None of the patients had any complications
which requiring a reoperation. Bleeding occurred in one case such as the Bilioma in one
case and Biliary fistula occurred in two cases.
The aim of this study was To evaluate the effect of gunshot injuries in maxillofacial region
in the context of Syrian crisis in terms of the quality of modern weapons which had used
and the clinical demonstration of these injuries (soft and bone),
in addition to evaluating
the quality of surgical treatment which had been provided to patients in the light of the
need for effective and successful treatments.
The research sample consisted of 40 patients with missile injuries who had reviewed:
Tishreen University Hospital- Alassad Hospital
Patients were divided by the area of injury (Mandible-Midface) and determined the
characteristics of each injury, All of the patients were managed by primary repair in the
majority of soft tissue defects. Bone repair was done primarily at the same stage using
different shapes of metal stabilizers.
In WDM networks, the end users exchange information with each other through all optical WDM
channels, called light-paths. A light-path must occupy the same wavelength on all the fiber links
through which it traverses. In a WDM optical network, with
a given set of connections, the
question of setting up light-paths by routing and allocating a wavelength to each connection is
called Routing and Wavelength Assignment (RWA) problem. Integer Linear Programming (ILP) is
a mathematical formulation helps in minimizing and maximizing an object function under multiple
constraints. This can help in formulating a mathematical model for RWA. This paper studies a
mathematical model for RWA in WDM Optical networks which can lead to a good network
planning. This paper suggests an enhancement proposal for Syrian telecommunication optical
networks using ILP formulation.
