Automatic Text Summarization Using Customizable Fuzzy Features and Attention on the Context and Vocabulary

Sahba, Ramin; Ebadi, Nima; Jamshidi, Mo; Rad, Paul

doi:10.23919/wac.2018.8430483

Cited by 31 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If there is uncertainty in the source data, indicator (8) cannot be used explicitly, since it does not provide for working with fuzzy values [2,[4][5]11,13,15,18]. Complete mismatch can be expressed as follows:…”

Section: Fig 1 Methods Of Fuzzy Multiparametric Selection Immentioning

confidence: 99%

Multiparametric selection of distance learning tools using fuzzy approaches

et al. 2020

View full text Add to dashboard Cite

The changes in connection with the outbreak of the pandemic have made their own adjustments to the educational process, almost simultaneously changing the usual way of full-time education at all levels, from school to higher professional education. The methodology of multiparametric selection of tools for ensuring the necessary quality of training, implemented on the principles of System of Systems, allowing the University to choose the composition and configuration of these tools, and taking into account the fuzziness of the source data, is proposed. Description of IM as fuzzy vectors and a method of evaluating misalignments can be the basis for multi-parameter methods of their choice, providing automated selection of the base set support the educational process. Thus, we can say that the remote format of work, which has entered all levels of the educational environment, is fixed there, and can be adjusted and developed – adapted to the specific conditions of educational institutions.

show abstract

Section: Fig 1 Methods Of Fuzzy Multiparametric Selection Immentioning

confidence: 99%

Multiparametric selection of distance learning tools using fuzzy approaches

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Inspired by Liu's (2016) Spatio-Temporal LSTM model, we apply a similar model that is equipped with a spatio-temporal mechanism to recognize the performed action and learn the exclusive motion patterns of the action. In other words, our inspired LSTM model utilizes two attention mechanisms [56]: attention over the time frames, and attention over various keypoint coordinates. Such spatio-temporal attention helps the model to understand an action despite variation among individuals preforming the same action with a certain intensity index, such as walking fast or punching hard.…”

Section: Lstm With Spatio-temporal Attentionmentioning

confidence: 99%

Human Action Performance Using Deep Neuro-Fuzzy Recurrent Attention Model

et al. 2020

Self Cite

View full text Add to dashboard Cite

A great number of computer vision publications have focused on distinguishing between human action recognition and classification rather than the intensity of actions performed. Indexing the intensity which determines the performance of human actions is a challenging task due to the uncertainty and information deficiency that exists in the video inputs. To remedy this uncertainty, in this paper we coupled fuzzy logic rules with the neural-based action recognition model to rate the intensity of a human action as intense or mild. In our approach, we used a Spatio-Temporal LSTM to generate the weights of the fuzzy-logic model, and then demonstrate through experiments that indexing of the action intensity is possible. We analyzed the integrated model by applying it to videos of human actions with different action intensities and were able to achieve an accuracy of 89.16% on our intensity indexing generated dataset. The integrated model demonstrates the ability of a neuro-fuzzy inference module to effectively estimate the intensity index of human actions.

show abstract

“…Updating the position of wolves during the hunting process using equations (9-11), does not guarantee that the value of every direction does not exceed the bounds of the problem. As a result, after attaining the discrete values of each direction through rounding toward zero the updated values of positions obtained from equation (9)(10)(11), an additional step is needed to be considered to return back the exceeding values into the boundaries of the problem. In the case of SCOS problem, upper bound denoted by ub , equals to the number of available candidates for each subtask, and lower bound, denoted by lb , is equal to 1.…”

Section: Range Checkingmentioning

confidence: 99%

“…Step5: Update the position of wolves through hunting process using Equations (9)(10)(11) Step6: Discretize the elements of position vectors obtained in step5 ( 1Xt ) using the floor function Step7: Perform range checking for values obtained in Step6 and map out of bound values (if any) into the allowed space according to Equation 12Step8: Select c Pn  number of parents for crossover operation and m Pn  number of parents for mutation operator according to Equation 13, where c P and m P are probability of crossover and mutation, respectively Step9: Perform the two-point crossover on the selected parents Step10: Perform the swap mutation operator on the selected individuals for mutation Step11: Merge the populations created in Steps 9 and 10 with the original population obtained after Step7…”

Section: Mutation Operatormentioning

confidence: 99%

“…Cloud manufacturing (CMfg) has been proposed as the panacea to implement this groundbreaking idea in recent years. Nonetheless, to fully achieve this goal, there remains significant issues with regard to enabling technologies [3][4][5][6][7], trust evaluation [8], task and resource description [9,10], scheduling [11][12][13] and service composition, among which the last one, is the key to offering composite services through the above-mentioned collaboration among partners. This has to be done by selecting the optimal resource for each sub-task out of the existing candidates at the resources pool and composing them to achieve the highest quality of service (QoS).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimal composition of tasks in cloud manufacturing platform: a novel hybrid GWO-GA approach

Bouzary¹,

Chen²,

Shahin³

2019

Preprint

View full text Add to dashboard Cite

Emerging cloud manufacturing paradigm aims towards providing highly integrated solutions via enabling cooperation between distributed manufacturing resources and capabilities. Implementation of this groundbreaking idea, however, is facing serious challenges springing from the currently centralized industrial structures. In order to help it make its way out, researchers have to address a number of pivotal issues in this regard. Service composition and optimal selection (SCOS), which tackles the problem of optimally selecting and combining available resources into a composite service is one of them. To deal with this NP-hard problem, we developed a novel hybrid algorithm based on the recentlyintroduced grey wolf optimizer (GWO) in which evolutionary operators are also embedded into the hunting mechanism of the basic algorithm.This approach not only makes it possible to adapt an algorithm with continuous structure such as GWO to solve a combinatorial problem such as SCOS, but also empowers it with providing higher exploration through crossover and mutation operators. Experiments conducted clearly proves the superior performance of the proposed algorithm over existing discrete variations of GWO and genetic algorithm, especially in large-scale SCOS problems.

show abstract

Automatic Text Summarization Using Customizable Fuzzy Features and Attention on the Context and Vocabulary

Cited by 31 publications

References 32 publications

Multiparametric selection of distance learning tools using fuzzy approaches

Multiparametric selection of distance learning tools using fuzzy approaches

Human Action Performance Using Deep Neuro-Fuzzy Recurrent Attention Model

Optimal composition of tasks in cloud manufacturing platform: a novel hybrid GWO-GA approach

Contact Info

Product

Resources

About