-
1
Solving aspect-based sentiment analysis task with GNN models and tree reconstruction methods
Published 2022“…Building the graph structure with dependency parsing algorithm and Tree-Lifting and Tree-Truncating algorithm. 2. Applying three different models: GCN, GAT, and Heterogeneous GCN on the data. 3. …”
Get full text
Thesis-Master by Coursework -
2
Boosting knowledge distillation and interpretability
Published 2021“…The generally accepted interpretable model is the decision tree. Although decision tree classification accuracy is not as good as deep neural networks, it is a more intuitive and interpretable common model. …”
Get full text
Thesis-Master by Coursework -
3
High-performance CMOS digital multiplier IC design
Published 2023“…Many new design ideas have emerged around how to improve the speed of multipliers, including pipeline structure, Wallace Tree structure, and Booth encoding. The Wallace Tree structure increases the speed of partial product summation by compressing the number of partial products; A Carry-look-ahead Adder can be greatly optimized for the addition of multi-bit numbers, and it skips the lengthy bit-by-bit carry propagation process, dramatically increasing the speed of addition operations. …”
Get full text
Thesis-Master by Coursework -
4
Design a 16-bit low power delay multiplier
Published 2021“…And in the architecture level, Dadda tree architecture tree is used.This kind of architecture can compress the accumulation process, reduce the delay time and save the number of adder cells. …”
Get full text
Thesis-Master by Coursework -
5
16-bit high speed CMOS multiplier IC design
Published 2021“…Then a 16-bit multiplier was designed based on the modified Booth 2 algorithm and Wallace tree technique, and the RTL code was implemented using Verilog HDL. …”
Get full text
Thesis-Master by Coursework -
6
16-bit low-power CMOS multiplier IC design
Published 2022“…In conclusion, the optimized Booth encoding Multiplier and Wallace Tree Multiplier were implemented with lower power consumption and shorter delay. …”
Get full text
Thesis-Master by Coursework -
7
High speed multiplier IC design based on booth algorithm
Published 2023“…It designed a high speed 32-bit signed multiplier based on Booth algorithm, 4-2 compressor using Wallace Tree structure and Carry-Look-Ahead adder. This design is 47. 03% faster than the original 32-bit signed shift multiplier, according to the length of critical path. …”
Get full text
Thesis-Master by Coursework -
8
SRAM-based in-memory computing for machine learning applications
Published 2022“…In this paper, a 4k weight-selective digital SRAM design is implemented with improvements on Bitcell and Adder Tree. It uses TG logic in the design to improve the speed and eliminate power consumption. …”
Get full text
Thesis-Master by Coursework -
9
Face swapping based on machine learning
Published 2021“…Firstly use Histogram of Oriented Gradient (HOG) to detect the face in a given image, and use training data to generate a prediction model based on the gradient boosting decision tree (GBDT) algorithm to extract the coordinates of 81 feature points of facial features and facial contours; Then train the Multi-Layer Perceptron (MLP) classifier to predict the gender and race of the face to be recognized and find the reference face image of the same gender race. …”
Get full text
Thesis-Master by Coursework -
10
Machine learning-based online stability assessment of power systems
Published 2021“…This dissertation uses three main machine learning methods, including DT (decision tree), RF (random forest), and SVM (support vector machine). …”
Get full text
Thesis-Master by Coursework -
11
Data mining and analysis of lift maintenance records
Published 2023“…Therefore, to predict the specific location and fault category of the lift failure on a certain day, this dissertation uses the Autoregressive Integrated Moving Average Model(AMIRA) method to predict the lift failure rate after the seasonal adjustment of the fault data.Combined with the chi-squared automatic interaction detector (CHAID) decision tree algorithm, the lift fault position and the lift fault type are successfully predicted. …”
Get full text
Thesis-Master by Coursework -
12
Low power-delay-product (PDP) CMOS multiplier design
Published 2021“…In architecture level, according to the result of modified Booth algorithm, a special Dadda tree is designed for partial product accumulation. …”
Get full text
Thesis-Master by Coursework -
13
Model development and simulation for robot-based warehouse automation
Published 2023“…The high level performs conflict search and constraint addition based on a binary tree. Considering that some of the commonly used warehouse layouts have oblique main aisles, the low level is allowed to adopt hybrid search methods to improve the efficiency of the system. …”
Get full text
Thesis-Master by Coursework -
14
Machine learning for mixed data
Published 2023“…These models include Logistic Regression (LR), K-Nearest Neighbors (KNN), Random Forest (RF), Gradient Boosting Decision Tree (GBDT) and XGBoost. We will collect the performance metrics data of each model at the optimal result. …”
Get full text
Thesis-Master by Coursework -
15
Predicting online video popularity using machine learning
Published 2024“…This study predicts online video popularity using different machine learning techniques such as Random Forest, Decision Tree, and XGBOOST, after comparing their prediction accuracy, standard deviation, and other parameters, I find that the XGBOOST model has the best prediction results, and I improve the Tuned XGBOOST model to get better results. …”
Get full text
Thesis-Master by Coursework -
16
Sampling-based robot path planning algorithms assessment and improvement
Published 2021“…Sampling-based planning algorithms (typically the RRT* family) represent one of the most popular path finding approaches that basically grow a tree or graph structure incrementally through various sampling techniques and try to connect the vertices with collision-free edges. …”
Get full text
Thesis-Master by Coursework -
17
Computational intelligence algorithm for indoor lighting control systems
Published 2020“…Analysis of the results revealed that Decision Tree algorithm has the best performance with an F1 score of 0.9072. …”
Get full text
Thesis-Master by Coursework -
18
Ultrahigh-resolution spectral CCD imaging based on machine learning
Published 2021“…Three advanced modeling methods were employed based on the Python library scikit-learn, namely, decision tree, k-nearest neighbor (KNN) and gradient boosting regression (GBR) model. …”
Get full text
Thesis-Master by Coursework -
19
Using AI and big data to predict stock market
Published 2023“…The prediction experiments are conducted by building and training three machine learning models: Decision Tree, Support Vector Machine, and deep learning model Long Short-Term Memory. …”
Get full text
Thesis-Master by Coursework -
20
Robost SLAM for complex scenes based on RADAR and LIDAR fusion
Published 2024“…In order to minimise the effect of this error, the initial point cloud data is first processed and reduced by voxel grid sampling, then the processed point cloud data is subjected to KD tree nearest neighbour search with suitable processing method, which gives the result of calculated normal vector and a satisfied evaluation of the quality of plane fitting. …”
Get full text
Thesis-Master by Coursework