Focused on the issue that existing group recommendation methods take less account of the implicit estimation of socialization relationships among group members and the use of group consensus to reduce the influence of preference conflicts, a Group Recommendation method based on implicit Trust and group Consensus (GR-TC) was proposed. The method was divided into a recommendation phase and a consensus phase. In the recommendation phase, implicit trust values were mined based on preference information and social relationships among members. The members’ individual preferences and weights, and the initial group preferences were estimated. In the consensus phase, inconsistent members were identified by consensus measurement and identification rules, a maximum harmony optimization consensus model was built, and the group recommendation list was obtained by adjusting and updating the group preferences. Experimental results show that social relationships among members affect group recommendation results, reasonable selection of implicit trust weights improves the harmony of inconsistent members. Compared with the traditional consensus feedback mechanism, the implicit trust-induced maximum harmony consensus feedback mechanism has less adjustment cost and less impact on inconsistent members.
Unsupervised relation extraction aims to extract the semantic relations between entities from unlabeled natural language text. Currently, unsupervised relation extraction models based on Variational Auto-Encoder (VAE) architecture provide supervised signals to train model through reconstruction loss, which offers a new idea to complete unsupervised relation extraction tasks. Focusing on the issue that this kind of models cannot understand contextual information effectively and relies on dataset inductive biases, a Prompt-based learning based Unsupervised Relation Extraction (PURE) model was proposed, including a relation extraction module and a link prediction module. In the relation extraction module, a context-aware Prompt template function was designed to fuse the contextual information, and the unsupervised relation extraction task was converted into a mask prediction task, so as to make full use of the knowledge obtained during pre-training phase to extract relations. In the link prediction module, supervised signals were provided for the relation extraction module by predicting the missing entities in the triples to assist model training. Extensive experiments on two public real-world relation extraction datasets were carried out. The results show that PURE model can use contextual information effectively and does not rely on dataset inductive biases, and has the evaluation index B-cubed F1 improved by 3.3 percentage points on NYT dataset compared with the state-of-the-art VAE architecture-based model UREVA (Variational Autoencoder-based Unsupervised Relation Extraction model).
In order to solve the problem of insufficient mining of potential association between remote nodes in human action recognition tasks, and the problem of high training cost caused by using multi-modal data, a multi-scale feature fusion human action recognition method under the condition of single mode was proposed. Firstly, the global feature correlation of the original skeleton diagram of human body was carried out, and the coarse-scale global features were used to capture the connections between the remote nodes. Secondly, the global feature correlation graph was divided locally to obtain the Complementary Subgraphs with Global Features (CSGFs), the fine-scale features were used to establish the strong correlation, and the multi-scale feature complementarity was formed. Finally, the CSGFs were input into the spatial-temporal Graph Convolutional module for feature extraction, and the extracted results were aggregated to output the final classification results. Experimental results show that the accuracy of the proposed method on the authoritative action recognition dataset NTU RGB+D60 is 89.0% (X-sub) and 94.2% (X-view) respectively. On the challenging large-scale dataset NTU RGB+D120, the accuracy of the proposed method is 83.3% (X-sub) and 85.0% (X-setup) respectively, which is 1.4 and 0.9 percentage points higher than that of the ST-TR (Spatial-Temporal TRansformer) under single modal respectively, and 4.1 and 3.5 percentage points higher than that of the lightweight SGN (Semantics-Guided Network). It can be seen that the proposed method can fully exploit the synergistic complementarity of multi-scale features, and effectively improve the recognition accuracy and training efficiency of the model under the condition of single modal.
To make use of the social relationship between experts and to consider the limited rationality of decision-making experts in group decision-making, a TODIM (TOmada de Decis?o Interativa Multicritério) group decision-making method under trust network was proposed. Firstly, according to the number of discussions of the experts, in each discussion, each expert would refer to his/her trustee’s decision matrix according to the degree of trust acceptance, and the decision matrices would be modified through information interaction and negotiation. Then, when the set number of expert discussions was met, the final group decision-making matrix was calculated. Finally, the TODIM group decision-making method under trust network and TODIM group decision-making method were applied to calculate the ranking results of different schemes. The ranking results were compared and analyzed, and the sensitivity analysis was performed on the number of expert discussions and trust acceptance. The case analysis results show that the TODIM group decision-making method under trust network can fully integrate trust network, ensure the multi-stage information interaction and feedback process in the decision-making process, and is superior to the general TODIM group decision-making method in comparison analysis and sensitivity analysis.
Refined short-term traffic flow prediction is the premise to ensure the rational decision making in Intelligent Transportation System (ITS). In order to establish the lane-changing model of self-driving car, predict vehicle trajectories, and guide vehicle routes, the timely traffic flow prediction for each lane has become an urgent problem to solve. However, refined short-term traffic flow prediction faces the following challenges: first, with the increasing diversity of traffic flow data, the traditional prediction methods cannot meet the requirements of ITS for high precision and short time delay; second, training prediction model for each lane make a huge waste of resources. To solve the above problems, a refined short-term traffic flow prediction model combined Convolutional-Gated Recurrent Unit (Conv-GRU) with Grey Relational Analysis (GRA) was proposed to predict lane flow. Considering the characteristics of long training time and relatively short reasoning time of deep learning, a cloud-fog deployment scheme was designed. Meanwhile, to avoid training prediction models for each lane, a model migration deployment scheme was proposed, which only needs to train the prediction model of some lanes, and then the trained prediction models were migrated to the associated lane for prediction through GRA. Experimental results of extensive comparisons on a real-world dataset show that, compared with traditional deep learning prediction methods, the proposed model has more accurate prediction performance; compared with Convolutional-Long Short-Term Memory (Conv-LSTM) network, the model has shorter running time. Furthermore, the model migration is realized by the proposed model under the condition of ensuring high-precision prediction, which saves about 49% of training time compared to training prediction model for each lane.
Traditional hypernetwork model is biased towards the majority class, which leads to much higher accuracy on majority class than the minority when being tackled on imbalanced data classification problem. In this paper, a Boosting ensemble of cost-sensitive hypernetworks was proposed. Firstly, the cost-sensitive learning was introduced to hypernetwork model, to propose cost-sensitive hyperenetwork model. Meanwhile, to make the algorithm adapt to the cost of misclassification on positive class, cost-sensitive hypernetworks were integrated by Boosting. The proposed model revised the bias towards the majority class when traditional hypernetwork model was tackled on imbalanced data classification, and improved the classification accuracy on minority class. The experimental results show that the proposed scheme has advantages in imbalanced data classification.
The seizure detection is important for the localization and classification of epileptic seizures. In order to solve the problem brought by large amount of data and high feature space in EEG (Electroencephalograph) for quickly and accurately detecting the seizures, a method based on max-Relevance and Min-Redundancy (mRMR) criteria and Extreme Learning Machine (ELM) was proposed. The time-frequency measures by Short-Time Fourier Transform (STFT) were extracted as features, and the large set of features were selected based on max-relevance and min-redundancy criteria. The states were classified using the extreme learning machine, Support Vector Machine (SVM) and Back Propagation (BP) algorithm. The result shows that the performance of ELM is better than SVM and BP algorithms in terms of computation time and classification accuracy. The classification accuracy rate of interictal durations and seizures can reach more than 98%, and the computation efficiency is only 0.8s. This approach can detect epileptic seizures accurately in real-time.
Concerning the problem that Fireworks Explosion Optimization (FEO) algorithm is easy to be premature and has low solution precision, an elite Opposition-Based Learning (OBL) was proposed. In every iteration, OBL was executed by the current best individual to generate an opposition search populations in its dynamic search boundaries, thus the search space of the algorithm was guided to approximate the optimum space. This mechanism is helpful to improve the balance and exploring ability of the FEO. For keeping the diversity of population, the sudden jump probability of the individual to the current best individual was calculated, and based on it, the roulette mechanism was adopted to choose the individual which entered into the child population. The experimental simulation on five classical benchmark functions show that, compared with the related algorithm, the improved algorithm has higher convergence rate and accuracy for numerical optimization, and it is suitable to solve the high dimensional optimization problem.