Many Distributed Denial of Service （DDoS） attack detection methods focus on improving model performance， but ignore the influence of traffic sample distribution and feature dimension on detection performance， resulting in the model learning redundant information. To address the problems of network traffic class imbalance and feature redundancy， a Hybrid Feature Selection method based on Multiple Evaluation Criteria （HFS-MEC） was proposed. Firstly， the Pearson Correlation Coefficient （PCC） and Mutual Information （MI） were considered comprehensively to select the correlation features. Then， the Sequential Backward Selection （SBS） algorithm based on Variance Inflation Factor （VIF） was designed to reduce the feature redundancy and further reduce the feature dimension. At the same time， to balance the detection performance and computation time， a Low-latency DDoS attack detection model based on Simple Recurrent Unit （SRU） （L-DDoS-SRU） was designed. Experiments were carried out on the CICIDS2017 and CICDDoS2019 datasets. The results show that HFS-MEC reduces the feature dimensions from 78 and 88 to 31 and 41， respectively； on the CICDDoS2019 dataset， L-DDoS-SRU reduces the detection time to only 40.34 seconds with a recall of 99.38%， which is improved by 8.47% compared to that of Long Short-Term Memory （LSTM）， and is increased by 9.76% compared to that of Gated Recurrent Unit （GRU）. The above verifies that the proposed method improves the detection performance and reduces the detection time effectively.

Searching and reusing relevant code can significantly improve software development efficiency. The deep learning-based code search models usually embed code pieces and query statements into the same vector space and then match and output the relevant code by computing cosine similarity； however， most of these models ignore the collaborative information between code pieces and query statements. To fully represent semantic information， a collaborative fusion-based code search model named BofeCS was proposed. Firstly， BERT （Bidirectional Encoder Representations from Transformers） model was utilized to extract the semantic information of the input sequences and then represent it as vectors. Secondly， a collaborative fusion network was constructed to extract the token-level collaborative information between code pieces and query statements. Finally， a residual network was built to alleviate the semantic information loss during the representation process. The multi-lingual dataset CodeSearchNet was used to carry out experiments to evaluate the effectiveness of BofeCS. Experimental results show that BofeCS can significantly improve the accuracy of code search and outperform the baseline models， UNIF （embedding UNIFication）， TabCS （Two-stage Attention-Based model for Code Search）， and MRCS （Multimodal Representation for neural Code Search）， in Mean Reciprocal Rank （MRR）， Normalized Discounted Cumulative Gain （NDCG）， and Top k Success hit Rate （SR@k）， where the MRR values are improved by 95.94%， 52.32%， and 16.95%， respectively.