A Novel Sentence Completion System for Punjabi using Deep Neural Networks

2022 ◽  
Vol 10 (1) ◽  
pp. 0-0

Sentence completion systems are actively studied by many researchers which ultimately results in the reduction of cognitive effort and enhancement in user-experience. The review of the literature reveals that most of the work in the said area is in English and limited effort spent on other languages, especially vernacular languages. This work aims to develop state-of-the-art sentence completion system for the Punjabi language, which is the 10th most spoken language in the world. The presented work is an outcome of the results of the experimentation on various neural network language model combinations. A new Sentence Search Algorithm (SSA) and patching system are developed to search, complete and rank the completed sub-string and give a syntactically rich sentence(s). The quantitative and qualitative evaluation metrics were utilized to evaluate the system. The results are quite promising, and the best performing model is capable of completing a given sub-string with more acceptability. Best performing model is utilized for developing the user-interface.

2020 ◽  
Vol 10 (4) ◽  
pp. 1340
Author(s):  
Heewoong Park ◽  
Jonghun Park

The task of sentence completion, which aims to infer the missing text of a given sentence, was carried out to assess the reading comprehension level of machines as well as humans. In this work, we conducted a comprehensive study of various approaches for the sentence completion based on neural language models, which have been advanced in recent years. First, we revisited the recurrent neural network language model (RNN LM), achieving highly competitive results with an appropriate network structure and hyper-parameters. This paper presents a bidirectional version of RNN LM, which surpassed the previous best results on Microsoft Research (MSR) Sentence Completion Challenge and the Scholastic Aptitude Test (SAT) sentence completion questions. In parallel with directly applying RNN LM to sentence completion, we also employed a supervised learning framework that fine-tunes a large pre-trained transformer-based LM with a few sentence-completion examples. By fine-tuning a pre-trained BERT model, this work established state-of-the-art results on the MSR and SAT sets. Furthermore, we performed similar experimentation on newly collected cloze-style questions in the Korean language. The experimental results reveal that simply applying the multilingual BERT models for the Korean dataset was not satisfactory, which leaves room for further research.


2019 ◽  
Vol 56 (4) ◽  
pp. 557-580 ◽  
Author(s):  
Sebastian Gabel ◽  
Daniel Guhl ◽  
Daniel Klapper

The authors propose a new, exploratory approach for analyzing market structures that leverages two recent methodological advances in natural language processing and machine learning. They customize a neural network language model to derive latent product attributes by analyzing the co-occurrences of products in shopping baskets. Applying dimensionality reduction to the latent attributes yields a two-dimensional product map. This method is well-suited to retailers because it relies on data that are readily available from their checkout systems and facilitates their analyses of cross-category product complementarity, in addition to within-category substitution. The approach has high usability because it is automated, is scalable and does not require a priori assumptions. Its results are easy to interpret and update as new market basket data are collected. The authors validate their approach both by conducting an extensive simulation study and by comparing their results with those of state-of-the-art, econometric methods for modeling product relationships. The application of this approach using data collected at a leading German grocery retailer underlines its usefulness and provides novel findings that are relevant to assortment-related decisions.


Author(s):  
Manjunath K. E. ◽  
Srinivasa Raghavan K. M. ◽  
K. Sreenivasa Rao ◽  
Dinesh Babu Jayagopi ◽  
V. Ramasubramanian

In this study, we evaluate and compare two different approaches for multilingual phone recognition in code-switched and non-code-switched scenarios. First approach is a front-end Language Identification (LID)-switched to a monolingual phone recognizer (LID-Mono), trained individually on each of the languages present in multilingual dataset. In the second approach, a common multilingual phone-set derived from the International Phonetic Alphabet (IPA) transcription of the multilingual dataset is used to develop a Multilingual Phone Recognition System (Multi-PRS). The bilingual code-switching experiments are conducted using Kannada and Urdu languages. In the first approach, LID is performed using the state-of-the-art i-vectors. Both monolingual and multilingual phone recognition systems are trained using Deep Neural Networks. The performance of LID-Mono and Multi-PRS approaches are compared and analysed in detail. It is found that the performance of Multi-PRS approach is superior compared to more conventional LID-Mono approach in both code-switched and non-code-switched scenarios. For code-switched speech, the effect of length of segments (that are used to perform LID) on the performance of LID-Mono system is studied by varying the window size from 500 ms to 5.0 s, and full utterance. The LID-Mono approach heavily depends on the accuracy of the LID system and the LID errors cannot be recovered. But, the Multi-PRS system by virtue of not having to do a front-end LID switching and designed based on the common multilingual phone-set derived from several languages, is not constrained by the accuracy of the LID system, and hence performs effectively on code-switched and non-code-switched speech, offering low Phone Error Rates than the LID-Mono system.


Algorithms ◽  
2021 ◽  
Vol 14 (2) ◽  
pp. 39
Author(s):  
Carlos Lassance ◽  
Vincent Gripon ◽  
Antonio Ortega

Deep Learning (DL) has attracted a lot of attention for its ability to reach state-of-the-art performance in many machine learning tasks. The core principle of DL methods consists of training composite architectures in an end-to-end fashion, where inputs are associated with outputs trained to optimize an objective function. Because of their compositional nature, DL architectures naturally exhibit several intermediate representations of the inputs, which belong to so-called latent spaces. When treated individually, these intermediate representations are most of the time unconstrained during the learning process, as it is unclear which properties should be favored. However, when processing a batch of inputs concurrently, the corresponding set of intermediate representations exhibit relations (what we call a geometry) on which desired properties can be sought. In this work, we show that it is possible to introduce constraints on these latent geometries to address various problems. In more detail, we propose to represent geometries by constructing similarity graphs from the intermediate representations obtained when processing a batch of inputs. By constraining these Latent Geometry Graphs (LGGs), we address the three following problems: (i) reproducing the behavior of a teacher architecture is achieved by mimicking its geometry, (ii) designing efficient embeddings for classification is achieved by targeting specific geometries, and (iii) robustness to deviations on inputs is achieved via enforcing smooth variation of geometry between consecutive latent spaces. Using standard vision benchmarks, we demonstrate the ability of the proposed geometry-based methods in solving the considered problems.


Electronics ◽  
2021 ◽  
Vol 10 (3) ◽  
pp. 230
Author(s):  
Jaechan Cho ◽  
Yongchul Jung ◽  
Seongjoo Lee ◽  
Yunho Jung

Binary neural networks (BNNs) have attracted significant interest for the implementation of deep neural networks (DNNs) on resource-constrained edge devices, and various BNN accelerator architectures have been proposed to achieve higher efficiency. BNN accelerators can be divided into two categories: streaming and layer accelerators. Although streaming accelerators designed for a specific BNN network topology provide high throughput, they are infeasible for various sensor applications in edge AI because of their complexity and inflexibility. In contrast, layer accelerators with reasonable resources can support various network topologies, but they operate with the same parallelism for all the layers of the BNN, which degrades throughput performance at certain layers. To overcome this problem, we propose a BNN accelerator with adaptive parallelism that offers high throughput performance in all layers. The proposed accelerator analyzes target layer parameters and operates with optimal parallelism using reasonable resources. In addition, this architecture is able to fully compute all types of BNN layers thanks to its reconfigurability, and it can achieve a higher area–speed efficiency than existing accelerators. In performance evaluation using state-of-the-art BNN topologies, the designed BNN accelerator achieved an area–speed efficiency 9.69 times higher than previous FPGA implementations and 24% higher than existing VLSI implementations for BNNs.


Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2778 ◽  
Author(s):  
Mohsen Azimi ◽  
Armin Eslamlou ◽  
Gokhan Pekcan

Data-driven methods in structural health monitoring (SHM) is gaining popularity due to recent technological advancements in sensors, as well as high-speed internet and cloud-based computation. Since the introduction of deep learning (DL) in civil engineering, particularly in SHM, this emerging and promising tool has attracted significant attention among researchers. The main goal of this paper is to review the latest publications in SHM using emerging DL-based methods and provide readers with an overall understanding of various SHM applications. After a brief introduction, an overview of various DL methods (e.g., deep neural networks, transfer learning, etc.) is presented. The procedure and application of vibration-based, vision-based monitoring, along with some of the recent technologies used for SHM, such as sensors, unmanned aerial vehicles (UAVs), etc. are discussed. The review concludes with prospects and potential limitations of DL-based methods in SHM applications.


2006 ◽  
Vol 14 (2) ◽  
pp. 223-253 ◽  
Author(s):  
Frédéric Lardeux ◽  
Frédéric Saubion ◽  
Jin-Kao Hao

This paper presents GASAT, a hybrid algorithm for the satisfiability problem (SAT). The main feature of GASAT is that it includes a recombination stage based on a specific crossover and a tabu search stage. We have conducted experiments to evaluate the different components of GASAT and to compare its overall performance with state-of-the-art SAT algorithms. These experiments show that GASAT provides very competitive results.


Sign in / Sign up

Export Citation Format

Share Document