The DOI matches its Crossref record; 39 of the 40 references it cites were found in a registry or an index; none of the 10 sources its claims cite is checked for support yet.
LSTM-CNN Architecture for Human Activity Recognition
Open accessAbstract
In the past years, traditional pattern recognition methods have made great progress. However, these methods rely heavily on manual feature extraction, which may hinder the generalization model performance. With the increasing popularity and success of deep learning methods, using these techniques to recognize human actions in mobile and wearable computing scenarios has attracted widespread attention. In this paper, a deep neural network that combines convolutional layers with long short-term memory (LSTM) was proposed. This model could extract activity features automatically and classify them with a few model parameters. LSTM is a variant of the recurrent neural network (RNN), which is more suitable for processing temporal sequences. In the proposed architecture, the raw data collected by mobile sensors was fed into a two-layer LSTM followed by convolutional layers. In addition, a global average pooling layer (GAP) was applied to replace the fully connected layer after convolution for reducing model parameters. Moreover, a batch normalization layer (BN) was added after the GAP layer to speed up the convergence, and obvious results were achieved. The model performance was evaluated on three public datasets (UCI, WISDM, and OPPORTUNITY). Finally, the overall accuracy of the model in the UCI-HAR dataset is 95.78%, in the WISDM dataset is 95.85%, and in the OPPORTUNITY dataset is 92.63%. The results show that the proposed model has higher robustness and better activity detection capability than some of the reported results. It can not only adaptively extract activity features, but also has fewer parameters and higher accuracy.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
The paper proposes an LSTM-CNN architecture that feeds raw mobile sensor data into a two-layer LSTM followed by convolutional layers, with global average pooling and batch normalization to reduce parameters and speed convergence. It reports high accuracy/F1 on UCI-HAR, WISDM, and OPPORTUNITY and compares favorably with CNN and DeepConvLSTM baselines.
Key findings
- On three public datasets, reported overall accuracy/F1 is UCI-HAR 95.78% (abstract/conclusion; experimental section says 95.80% accuracy), WISDM 95.85% (abstract/conclusion; experimental section says 95.75% accuracy), and OPPORTUNITY 92.63%; without the Null class, OPPORTUNITY accuracy is 87.58%.
- Compared with the CNN of Yang et al. and DeepConvLSTM under the same scenario, LSTM-CNN increased about 7% on OPPORTUNITY and averaged 3% on UCI-HAR and WISDM by F1 score.
- Replacing the fully connected layer with GAP reduced parameters: model B has 27,462 parameters versus more than 502 thousand for model A, about 94% less, with almost the same performance; adding BN raised F1 to 93.35%, and adding LSTM layers raised UCI-HAR F1 to 95.78%.
- Hyperparameter tests on UCI-HAR found Adam was the best optimizer, batch size 192 gave the highest accuracy, and 192 filters in the second convolutional layer gave F1 96.38% (4% over 64 filters) but increased model parameters by more than 70%.
Methods
- Architecture: preprocessed sensor data passes through two LSTM layers (32 memory cells each, 64 total), then two convolutional layers (64 filters of size 1×5 with stride 2; 128 filters of size 1×3 with stride 1), with max-pooling between them, followed by GAP, BN, and a dense Softmax output layer.
- Data: UCI-HAR (30 subjects, 6 activities, 50 Hz, 748,406 samples), WISDM (36 subjects, 6 activities, 20 Hz, 1,098,209 samples, unbalanced), and OPPORTUNITY (4 subjects, 17 activities plus Null class, 30 Hz, 113 channels). Preprocessing used linear interpolation, 0–1 normalization, and sliding windows of length 128 for UCI/WISDM and 24 for OPPORTUNITY, with 50% overlap.
- Training/evaluation: Keras with TensorFlow backend, cross-entropy loss, Adam optimizer, batch size 192, 200 epochs, learning rate 0.001; evaluated with accuracy and F1 score, confusion matrices, and comparisons to CNN and DeepConvLSTM.
Limitations
- The paper states WISDM and OPPORTUNITY are imbalanced, so overall accuracy is not an appropriate measure and F1 is used; it also notes relatively poor discrimination between sitting and standing in UCI-HAR, with recall and precision in the range of 92%–93%.
- The paper states LSTM layers slow computation because each time step depends on the previous one and cannot be computed in parallel; LSTM-CNN took 9,416 ms per epoch versus 5,156 ms for model D, and GAP can shift training pressure to convolutional layers and slow convergence, which BN was added to address.
- Evident from the text: reported metrics are inconsistent across sections (e.g., UCI-HAR 95.78% vs 95.80%; WISDM 95.85% vs 95.75%), and OPPORTUNITY evaluation uses only four subjects, limiting generalization claims.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 of 10 cited sources checked · 0 supported · 0 flagged
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, the overall accuracy of the model in the UCI-HAR dataset is 95.78%, in the WISDM dataset is 95.85%, and in the OPPORTUNITY dataset is 92.63%.”
§ LSTM-CNN Architecture for Human Activity Recognition > Abstract
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, the overall accuracy of the model in the UCI-HAR dataset is 95.78%, in the WISDM dataset is 95.85%, and in the OPPORTUNITY dataset is 92.63%.”
§ LSTM-CNN Architecture for Human Activity Recognition > Abstract
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, the overall accuracy of the model in the UCI-HAR dataset is 95.78%, in the WISDM dataset is 95.85%, and in the OPPORTUNITY dataset is 92.63%.”
§ LSTM-CNN Architecture for Human Activity Recognition > Abstract
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, the model parameters increase by more than 70%.”
§ LSTM-CNN Architecture for Human Activity Recognition > 2) EFFECT OF NUMBER OF FILTERS
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, an overall accuracy of 92.63% was achieved.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In addition, when the Null class is removed from the classification task (see Table 10), our method achieved an overall accuracy of 87.58% in the gesture recognition task.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can be seen that the accuracy reaches the highest when the batch size is selected as 192.”
§ LSTM-CNN Architecture for Human Activity Recognition > 3) EFFECT OF BATCH SIZE
Strength:plain· not yet calibratedCausal:correlational· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, the F 1 score of this model reaches 93.35%.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesThe paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Finally, it outperforms the model C by 1% on average.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesThe paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“There was relatively poor discrimination between sitting and standing.The recall and precision were in the range of 92%∼93%.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes
- It may state two separate claims as one. 28
- Probably not a substantial claim. 10
- The quote only partly states this claim. 1
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Accordingly, there are more than 502 thousand model parameters, of which the parameters of fully-connected layers occupy the most.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesIt may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“Compared with the CNN model of Yang et al., LSTM-CNN has a significant increase of about 7% for the OPPORTUNITY dataset and is superior to the DeepConvLSTM model.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“F 1 score reaches 96.38% when the number of filters is selected as 192, which outperforms when the number of filters is 64 by 4%.”
§ LSTM-CNN Architecture for Human Activity Recognition > 2) EFFECT OF NUMBER OF FILTERS
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“And it takes 1681 milliseconds per epoch in the training stage.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The parameters of the model B are only 27462, which is about 94% less than that of the model A while the performance remains almost the same.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“It should be noted that the model parameters have been greatly reduced under adding the GAP layer to the network.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✗ Based: failsProbably not a substantial claim. The paper may give no reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“They are 5156 milliseconds per epoch and 9416 milliseconds per epoch, respectively.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can be seen that the model trained by Adam optimizer has the best fitting effect and the fluctuation of gradient descent curve is the most stable.”
§ LSTM-CNN Architecture for Human Activity Recognition > 1) EFFECT OF OPTIMIZER
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Eventually, the F 1 score reached 95.78%, 95.85% and 92.63% on the UCI-HAR, WISDM and OPPORTUNITY datasets, respectively.”
§ LSTM-CNN Architecture for Human Activity Recognition > VI. CONCLUSION
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can also be seen that LSTM-CNN outperforms the other two models on the UCI-HAR and WISDM datasets, with the best-reported result increasing by an average of 3%.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For the UCI-HAR dataset, there were 2940 instances that have been correctly classified, and the overall accuracy reached 95.80%.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The main reason may be that the two activities are similar from the perspective of motion sensors.It is difficult to mine deeper information only by acceleration and angular velocity data.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. EVALUATION ON THREE PUBLIC DATASETS
Strength:hedged· not yet calibratedCausal:conditional· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“To address this problem, researchers have turned to deep learning methods that could automatically extract appropriate features from raw sensor data during the training phase and present the low-level original temporal features with high-level abstract sequences.”
§ LSTM-CNN Architecture for Human Activity Recognition > II. RELATED WORK
What the sources say
- Not checked yetcites [24]- [27]
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Although the above models could generally recognize human activities, the overall network structure is relatively complex.In addition, these models have a large number of parameters, which results in high computational cost.It is difficult to be used in occasions that require high real-time performance.”
§ LSTM-CNN Architecture for Human Activity Recognition > II. RELATED WORK
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“To sum up, compared with the methods proposed in other literatures, the LSTM-CNN model shows consistent superior performance and has good generalization.”
§ LSTM-CNN Architecture for Human Activity Recognition > VI. CONCLUSION
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The model proposed in this paper not only could achieve high recognition accuracy but also greatly simplify the model structure.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The model C adds a BN layer after the GAP layer to stabilize the output of the upper layer.It speeds up the convergence of the model and improves accuracy.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Furthermore, GAP sums out the spatial information, so it is more robust to the spatial transformation of the input.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. GLOBAL AVERAGE POOLING LAYER
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✗ Based: failsProbably not a substantial claim. The paper may give no reason why it holds. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“There is no parameter to optimize in the GAP layer.Thus, it achieves the goal of reducing global model parameters.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. GLOBAL AVERAGE POOLING LAYER
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Among them, the recordings of 22 subjects were used to build the training set, and the rest was used to build the test set.”
§ LSTM-CNN Architecture for Human Activity Recognition > V. EXPERIMENTAL RESULTS
Strength:plain· not yet calibratedn = 22Tests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The training set is composed of the recordings of 30 subjects and the remaining recordings of 6 subjects are used to build the test set.The two parts are completely separate.”
§ LSTM-CNN Architecture for Human Activity Recognition > V. EXPERIMENTAL RESULTS
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can not only adaptively extract activity features, but also has fewer parameters and higher accuracy.”
- Also stated in: § II. RELATED WORK · plain
“The results show that the proposed model not only has high accuracy but also has good generalization ability and fast convergence speed.”
§ LSTM-CNN Architecture for Human Activity Recognition > Abstract
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesThe quote only partly states this claim.
- Also stated in: § II. RELATED WORK · plain
In the paper
“Although the proposed model has high accuracy and brief architecture, it was only evaluated on one dataset which has just six activities, which does not prove that the proposed model has good generalization ability.”
§ LSTM-CNN Architecture for Human Activity Recognition > II. RELATED WORK
What the sources say
- Not checked yetcites [28] source
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesIt may state two separate claims as one.
In the paper
“Machine learning methods may rely heavily on heuristic manual feature extraction in most daily human activity recognition tasks.It is usually limited by human domain knowledge [23].”
§ LSTM-CNN Architecture for Human Activity Recognition > II. RELATED WORK
What the sources say
- Not checked yetcites [23] source
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The BN layer normalizes and reconstructs the input data on each batch of training samples to ensure the stability of the output of the previous layer, so as to improve the training speed and accuracy.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. BATCH NORMALIZATION LAYER
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim. It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“During the training process, the distribution of input data of each layer would continuously change due to the weight parameters of the upper layer are constantly updated.Therefore, it is necessary to change the weight parameters to adapt to this new distribution, which leads to difficulty in network training and slows down the convergence speed.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. BATCH NORMALIZATION LAYER
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
In the paper
“CNN has gained increasing popularity because of its ability to learn unique representations from images or speech [34].”
§ LSTM-CNN Architecture for Human Activity Recognition > B. CONVOLUTIONAL AND POOLING LAYERS
What the sources say
- Not checked yetcites [34] source
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“To sums up, the strategy of using the global average pooling layer and batch normalization layer to replace the fully-connected layer is effective.Moreover, the method of using LSTM to extract the temporal information to improve model performance is favorable.”
§ LSTM-CNN Architecture for Human Activity Recognition > D. IMPACT OF NETWORK STRUCTURE ON MODEL PERFORMANCE
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
- Rests on an assumption: Recordings of activities based on mobile sensors are temporal sequences.
- Rests on an assumption: LSTM has the ability to capture temporal information from sequential data.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Each node of the fully-connected layer is connected to the nodes of the upper layer so that the features extracted from the upper layer could be merged.It makes up for the shortcomings of the GAP layer in this regard.”
§ LSTM-CNN Architecture for Human Activity Recognition > E. OUTPUT LAYER
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✗ Based: failsProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds. It may state two separate claims as one.
In the paper
“The results showed an excellent effect, but the model converged hardly, causing a lot of time to be wasted in the training stage.”
§ LSTM-CNN Architecture for Human Activity Recognition > II. RELATED WORK
What the sources say
- Not checked yetcites [29] source
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“For instance, in the model Krizhevsky [35], the first fully-connected layer FC1 has 4096 nodes, and the output of the upper pooling layer Max-Pool3 has 9216 nodes.Thus, there would be more than 37 million weight parameters between the MaxPool3 layer and the FC1 layer, which would consume a lot of memory and computational cost.”
§ LSTM-CNN Architecture for Human Activity Recognition > C. GLOBAL AVERAGE POOLING LAYER
What the sources say
- Not checked yetcites [35]
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It can not only avoid complex feature extraction but also has high recognition accuracy under the premise of a few model parameters.”
§ LSTM-CNN Architecture for Human Activity Recognition > VI. CONCLUSION
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
The paper relies on this without support.
LSTM is a variant of RNN that is more suitable for processing temporal sequences.
“LSTM is a variant of the recurrent neural network (RNN), which is more suitable for processing temporal sequences.”
The paper relies on this without support.
LSTM can eliminate the gradient vanishing limitation of RNN.
“LSTM is a variety of RNN, which could eliminate this limitation.”
The paper relies on this without support.
LSTM has great advantages in feature extraction of sequence data than convolutional neural networks due to its special memory cells.
“LSTM has great advantages in feature extraction of sequence data than convolutional neural networks due to its special memory cells.”
The paper relies on this without support.
RNN can take advantage of the chronological relationship between sensor readings.
“RNN could take advantage of the chronological relationship between sensor readings.”
The paper relies on this without support.
RNN has the problem of gradient vanishing, which hinders the ability of the network to model between raw sensor data and human activities in a long context window.
“Although RNN has the ability to capture temporal information from sequential data, it has the problem of gradient vanishing, which hinders the ability of the network to model between raw sensor data and human activities in a long context window.”
The paper relies on this without support.
The more convolution kernels are used, the more hidden features could be mined in the input samples.
“Generally speaking, the more convolution kernels are used, the more hidden features could be mined in the input samples.”
The paper relies on this without support.
The max-pooling layer filters interference noise caused by the unconscious jitter of the human body.
“It serves two purposes.One is to reduce the parameters while maintaining dominant features, the other is to filter the interference noise caused by the unconscious jitter of the human body.”
The paper relies on this without support.
The WISDM and OPPORTUNITY datasets are imbalanced.
“The WISDM and OPPORTU-NITY mentioned above are both imbalanced datasets.”
The paper relies on this without support.
Hyper-parameters have a great impact on classification model performance.
“Hyper-parameters have a great impact on the classification model performance.”
The paper relies on this without support.
The weight parameters of CNN mainly concentrate on the fully-connected layer.
“The weight parameters of CNN mainly concentrate on the fully-connected layer.”
The paper relies on this without support.
Recordings of activities based on mobile sensors are temporal sequences.
“The recordings of activities based on mobile sensors are temporal sequences and LSTM has the ability to capture temporal information from sequential data.”
The paper relies on this without support.
LSTM has the ability to capture temporal information from sequential data.
“The recordings of activities based on mobile sensors are temporal sequences and LSTM has the ability to capture temporal information from sequential data.”
The paper relies on this without support.
When training LSTM layers, the calculation of each time step depends on the output of the previous time step.
“When training LSTM layers, the calculation of each time step depends on the output of the previous time step.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“In general, it is examined that the more convolution kernels are used, the more hidden features of the input samples could be extracted [48].”
From Deep CNN-LSTM With Self-Attention Model for Human Activity Recognition Using Wearable Sensor (2022) · cites this paper as [48]
In the citing paper
“In contrast to these three models, i.e.,sequential LSTM layers (LSTM-CNN) before Convolution layer [48], sequential Convolution and dropout layers (CNN-LSTM) before LSTM layer [59], parallel LSTM layers (Parrallel LSTM-CNN) with Convolution layer [60], our chosen model achieved the best testing result of 99.93% for H-Activity dataset and 98.76% and 93.11% for MHEALTH and UCI-HAR dataset respective”
From Deep CNN-LSTM With Self-Attention Model for Human Activity Recognition Using Wearable Sensor (2022) · cites this paper as [48]
In the citing paper
“Hence, all of the above-mentioned models [48], [59], [60] were implemented with exact parameters described in the respective literature except the input and output layer to match our dataset.”
From Deep CNN-LSTM With Self-Attention Model for Human Activity Recognition Using Wearable Sensor (2022) · cites this paper as [48]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
UCI-HAR overall accuracy has two reported values
MajorThe abstract reports 95.78% overall accuracy for UCI-HAR, whereas the results section reports 95.80% for the same measure.
“the overall accuracy of the model in the UCI-HAR dataset is 95.78%”
“For the UCI-HAR dataset, there were 2940 instances that have been correctly classified, and the overall accuracy reached 95.80%.”
WISDM waveform duration conflicts with its sampling rate
MinorAt the stated WISDM sampling rate of 20 Hz, 128 points span 6.4 seconds, not 2.56 seconds.
Reported 2.56 · recomputed 6.4
“Fig. 1 shows the acceleration waveform of 2.56 seconds (128 points in total) of each activity”
UCI-HAR is incorrectly described as having the most volunteers
MinorThe paper says UCI-HAR has the largest number of volunteers at 30, but states that WISDM has 36 subjects.
“The UCI-HAR dataset has the largest number of volunteers, which means that this dataset was constructed from the recordings of 30 subjects.”
“Its experimental object consists of 36 subjects.”
Adding an LSTM layer to model D leaves the stated total unchanged
MinorModel D is described as having two LSTM layers. The paper then says it added another layer to model D for a total of two layers.
“In model D, the data captured from mobile sensors are firstly fed into two layers of LSTMs and then transmitted to convolutional layers for feature extraction.”
“In our work, we added another layer of LSTM, in a total of 2 layers, on the basis of model D to further improve the model performance.”
WISDM overall accuracy has two reported values
MinorThe abstract reports 95.85% overall accuracy for WISDM, whereas the results section reports 95.75%.
“in the WISDM dataset is 95.85%”
“the overall accuracy of the WISDM dataset (an unbalanced dataset) reached 95.75%.”
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Activity recognition using cell phone accelerometers
Kwapisz, J. R., Weiss, G. M., & Moore, S. A. (2011) · ACM SIGKDD Explorations Newsletter · doi:10.1145/1964897.1964918 · Cited by 2,912 · cited 1×
“The WISDM dataset [31] has a total of 1098209 samples, and the percentage of the total samples associated with each activity was shown in Table 3.”
BackgroundDeep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761 · cited 2×
“[25] proposed an activity recognition classi-fier, which combined deep CNN and LSTM to classify 27 hand gestures and five movements.”
BackgroundDeep learning for sensor-based activity recognition: A survey
Wang, J., Chen, Y., Hao, S., Peng, X., & Hu, L. (2019) · Pattern Recognition Letters · doi:10.1016/j.patrec.2018.02.010 · Cited by 2,014 · cited 1×
“Human activity recognition (HAR) plays an important role in people’s daily lives because it has the ability to learn profound advanced knowledge about human activities from raw sensor data [1].”
BackgroundHuman activity recognition with smartphone sensors using deep learning neural networks
Ronao, C. A., & Cho, S.-B. (2016) · Expert Systems with Applications · doi:10.1016/j.eswa.2016.04.032 · Cited by 1,215
Sensor-Based Activity Recognition
Chen, L., Hoey, J., Nugent, C. D., Cook, D. J., & Yu, Z. (2012) · IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) · doi:10.1109/tsmcc.2012.2198883 · Cited by 1,074
(2015) · Cited by 989
Collecting complex activity datasets in highly rich networked sensor environments
Roggen, D., Calatroni, A., Rossi, M., Holleczek, T., Forster, K., Troster, G., Lukowicz, P., Bannach, D., Pirkl, G., Ferscha, A., et al. (2010) · 2010 Seventh International Conference on Networked Sensing Systems (INSS) · doi:10.1109/inss.2010.5573462 · Cited by 788
The Opportunity challenge: A benchmark database for on-body sensor-based activity recognition
Chavarriaga, R., Sagha, H., Calatroni, A., Digumarti, S. T., Tröster, G., Millán, J. D. R., & Roggen, D. (2013) · Pattern Recognition Letters · doi:10.1016/j.patrec.2012.12.014 · Cited by 742 · cited 3×
“In terms of the sensor setting, theOPPORTUNITY challenge guidelines [33] were adopted.”
MethodTransition-Aware Human Activity Recognition Using Smartphones
Reyes-Ortiz, J.-L., Oneto, L., Samà, A., Parra, X., & Anguita, D. (2016) · Neurocomputing · doi:10.1016/j.neucom.2015.07.085 · Cited by 740
Time Series Classification Using Multi-Channels Deep Convolutional Neural Networks
Zheng, Y., Liu, Q., Chen, E., Ge, Y., & Zhao, J. L. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-08010-9_33 · Cited by 706
Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables
Hammerla, N. Y., Halloran, S., & Ploetz, T. (2016) · arXiv · doi:10.48550/arxiv.1604.08880 · Cited by 606
A Knowledge-Driven Approach to Activity Recognition in Smart Homes
Chen, L., Nugent, C., & Wang, H. (2011) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2011.51 · Cited by 591
Self-citationDeep Learning of Representations: Looking Forward
Bengio, Y. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-39593-2_1 · Cited by 568
Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates
Liu, J., Shahroudy, A., Xu, D., Kot, A. C., & Wang, G. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2771306 · Cited by 548
A Survey on Activity Detection and Classification Using Wearable Sensors
Cornacchia, M., Ozcan, K., Zheng, Y., & Velipasalar, S. (2016) · IEEE Sensors Journal · doi:10.1109/jsen.2016.2628346 · Cited by 427
Hand Gesture Recognition Using Micro-Doppler Signatures With Convolutional Neural Network
Kim, Y., & Toomajian, B. (2016) · IEEE Access · doi:10.1109/access.2016.2617282 · Cited by 384
Feature Selection and Activity Recognition System Using a Single Triaxial Accelerometer
Gupta, P., & Dallas, T. (2014) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2014.2307069 · Cited by 335
InnoHAR: A Deep Neural Network for Complex Human Activity Recognition
Xu, C., Chai, D., He, J., Zhang, X., & Duan, S. (2019) · IEEE Access · doi:10.1109/access.2018.2890675 · Cited by 333
BodyScope: a wearable acoustic sensor for activity recognition
Yatani, K., & Truong, K. N. (2012) · Proceedings of the 2012 ACM Conference on Ubiquitous Computing · doi:10.1145/2370216.2370269 · Cited by 310
Wearable Inertial Sensors for Human Motion Analysis: A Review
Lopez-Nava, I. H., & Munoz-Melendez, A. (2016) · IEEE Sensors Journal · doi:10.1109/jsen.2016.2609392 · Cited by 263
Wearable Sensor-Based Hand Gesture and Daily Activity Recognition for Robot-Assisted Living
Zhu, C., & Sheng, W. (2011) · IEEE Transactions on Systems Man and Cybernetics - Part A Systems and Humans · doi:10.1109/tsmca.2010.2093883 · Cited by 234
Fast unsupervised ego-action learning for first-person sports videos
Kitani, K., Okabe, T., Sato, Y., & Sugimoto, A. (2011) · doi:10.1109/cvpr.2011.5995406 · Cited by 223
Human Activity Classification in Smartphones Using Accelerometer and Gyroscope Sensors
Jain, A., & Kanhangad, V. (2017) · IEEE Sensors Journal · doi:10.1109/jsen.2017.2782492 · Cited by 223
A Lightweight Deep Learning Model for Human Activity Recognition on Edge Devices
Agarwal, P., & Alam, M. (2020) · Procedia Computer Science · doi:10.1016/j.procs.2020.03.289 · Cited by 186
Super Normal Vector for Human Activity Recognition with Depth Cameras
Yang, X., & Tian, Y. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2565479 · Cited by 162
A-Wristocracy: Deep learning on wrist-worn sensing for recognition of user complex activities
Vepakomma, P., De, D., Das, S. K., & Bhansali, S. (2015) · 2015 IEEE 12th International Conference on Wearable and Implantable Body Sensor Networks (BSN) · doi:10.1109/bsn.2015.7299406 · Cited by 136
Margarito, J., Helaoui, R., Bianchi, A., Sartor, F., & Bonomi, A. (2015) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2015.2471094 · Cited by 118
Recognizing Human Activity in Free-Living Using Multiple Body-Worn Accelerometers
Fullerton, E., Heller, B., & Munoz-Organero, M. (2017) · IEEE Sensors Journal · doi:10.1109/jsen.2017.2722105 · Cited by 81
Lin, Y., Le Kernec, J., Yang, S., Fioranelli, F., Romain, O., & Zhao, Z. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2872849 · Cited by 81
Sum Product Networks for Activity Recognition
Amer, M. R., & Todorović, S. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2465955 · Cited by 78
Compressive Sequential Learning for Action Similarity Labeling
Qin, J., Liu, L., Zhang, Z., Wang, Y., & Shao, L. (2016) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2508600 · Cited by 55
Muñoz-Organero, M. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2882943 · Cited by 51
Ronaoo, C. A., & Cho, S. (2015) · 한국정보과학회 학술발표논문집 · Cited by 43
Activity Recognition Using Complex Network Analysis
Jalloul, N., Porée, F., Viardot, G., Hostis, P. L., & Carrault, G. (2017) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2017.2762404 · Cited by 32
Concurrent Recognition of Cross-Scale Activities via Sensorless Sensing
Wang, L., Su, X., Jing, N., Wenyuan, L., & Binbin, L. (2018) · IEEE Sensors Journal · doi:10.1109/jsen.2018.2874506 · Cited by 14
Human activity recognition using a low cost, COTS radar network
Çağlıyan, B., Karabacak, C., & Gürbüz, S. Z. (2014) · Proceedings of the IEEE National Radar Conference · doi:10.1109/radar.2014.6875784 · Cited by 13
A Lightweight Deep Learning Model for Human Activity Recognition on Edge Devices
Agarwal, P., & Alam, M. (2019) · arXiv · doi:10.48550/arxiv.1909.12917
Flagged first, then by citations
1 of the 40 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Deep Learning in Human Activity Recognition with Wearable Sensors: A Review on Advances
Zhang, S., Li, Y., Zhang, S., Shahabi, F., Xia, S., Deng, Y., & Alshurafa, N. (2022) · Sensors · doi:10.3390/s22041476 · Cited by 625
“Similarly, other researchers [8, 82, 94, 95, 197, 245, 301, 302], [238] have also developed the CNN-LSTM model in varios application scenarios by taking advantage of the feature extraction ability of CNN and the time-series data reasoning ability of LSTM.”
BackgroundRamanujam, E., Perumal, T., & Padmavathi, S. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3069927 · Cited by 455
A review of machine learning in processing remote sensing data for mineral exploration
Shirmard, H., Farahbakhsh, E., Müller, R. D., & Chandra, R. (2021) · Remote Sensing of Environment · doi:10.1016/j.rse.2021.112750 · Cited by 448
Multi-input CNN-GRU based human activity recognition using wearable sensors
Dua, N., Singh, S. N., & Semwal, V. B. (2021) · Computing · doi:10.1007/s00607-021-00928-8 · Cited by 432
Materials-Driven Soft Wearable Bioelectronics for Connected Healthcare
Gong, S. P., Yan, L., Yin, J., Levin, A., & Cheng, W. (2024) · Chemical Reviews · doi:10.1021/acs.chemrev.3c00502 · Cited by 433
Metadata
- Authors
- Xia, K., Huang, J., & Wang, H.
- Affiliations
- University of Shanghai for Science and Technology (CN)
- Venue
- IEEE Access
- Published
- 2020
- Volume
- 8
- Pages
- 56855-56866
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3011785450
- Cited by
- 851
- References
- 40
- Citations by year
- 2026: 68 · 2025: 169 · 2024: 169 · 2023: 185 · 2022: 136 · 2021: 83 · 2020: 14
- FWCI
- 40.64
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Natural Science Foundation of China (512070901, 51637001)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Physical Sciences › Engineering › Biomedical Engineering › Gait Recognition and Analysis
- Keywords
- human activity recognition, long short-term memory, convolutional neural networks, mobile sensor data, wearable computing, automatic feature extraction, global average pooling, batch normalization, UCI HAR dataset, WISDM dataset, Opportunity dataset
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
