Traditional text representation techniques will produce high-dimensional and sparse vectors for short texts since they have fewer words, which increases the likelihood of noise interference and lowers the accuracy of short text clustering. An alternative model, CASA-Bert-K, is investigated to overcome the aforementioned issues. First, In order to add contextual features into the text representation, context-awareness self-attention networks are combined with Bert. Secondly, To de-optimize the representation of short texts, a simple and effective data augmentation approach to contrast learning is introduced. The text representation module and the clustering module are jointly trained to optimize both text representation and clustering in order to further improve the clustering accuracy. The experimental findings show the researched model's effectiveness in raising the accuracy rate through the use of three datasets in the tests.
In order to effectively monitor and detect the characteristics and abnormal behavior of network traffic, this paper uses convolution neural network model to realize the classification and prediction of network data. Firstly, a system model including data preprocessing module, data hierarchy module and convolution neural network module is established. Secondly, the data is preprocessed and normalized by convolution operation. Then, the pooled pool is used to select features, and the back-propagation algorithm and gradient descent algorithm are used to optimize the model to complete feature extraction and training. Based on this, the characteristics and abnormal behavior of network traffic can be monitored and detected. In the experiment, the KDDCUP 1999 data set and UNSW-NB15 data set were used for system testing and compared with various methods. The experimental results show that the detection accuracy, precision, recall and F1 index of the proposed system are more than 10% higher than those of other methods, showing good performance and high application value.
Using wireless body area networks to monitor and recognize human motion is one of the research hotspots in wireless body area networks, and gesture is an important component of human motion. Gesture, as a natural, intuitive, and easy to learn human-computer interaction method, can provide operational convenience for health monitoring in wireless body area networks. Gestures are divided into static and dynamic gestures. Static gestures rely solely on the external shape and contour information of the hand to convey information. Dynamic gestures also include gesture timing information for gesture movements, which can express richer and more accurate information. Data acquisition methods for gesture movements include data glove based, vision based, EMG based, and inertial sensor based methods. This paper studies a dynamic gesture recognition method based on acceleration sensors.
The research results of Chinese Medical academic articles are more accurate and representative than other medical texts. This paper studies the extraction method of medical entity recognition in Chinese academic articles on hypertension. This paper proposes a BiLSTM-CRF (Bidirectional Long Short-Term Memory and Conditional Random Fields) model blended with TFE (Text Feature Embedding) and BERT (Bidirectional Encoder Representation from Transformers) for Chinese medical entity recognition. BERT model was used for text pre-processing while text features of Chinese medical academic articles were obtained, and then the annotation and classification of medical entity were completed via three models. Compared with the BiLSTM-CRF model and BERT-BiLSTM-CRF model, the recognition effect of the BERT-TFE-BiLSTM-CRF model proposed in this study was better than the other two models and the F1 value of model was 6.9% and 3.3% higher respectively. The BERT-TFE-BiLSTM-CRF model can effectively recognize medical entity in Chinese academic articles on hypertension.
With the trend of globalization, modern information technology breathes new life into all walks of life. Concerning the complexity of English phonetic training, it’s urgent to explore a practical system related to information technology. Cloud service system provides a possible new educational technique, which would help English phonetic training achieve man-machine interaction, computing, Cloud storing, and integrating conveniently and efficiently. This paper introduces the background of English phonetic training under the circumstances of Cloud Platform. Then, current information technologies related to English phonetic training are exhibited. It proposes the feasible ways to construct English phonetic training system on Cloud Platform. By experiment, it makes quantitative and qualitative analysis, and further testifies the validity of this construction by analysing and discussing the results of the experiment due to statistical data. It draws the conclusion that to construct English phonetic training system on Cloud Platform is feasible and effective. It helps arouse more scholars to more concern about application of English phonetic training system on Cloud Platform.
Designed and implemented a wireless flexible wearable music controller which based on flexible fabric sensor and micro-control unit, controller can be integrated into clothing, provides a new idea for the preparation of a flexible wearable sensor system with simple structure and low cost. The controller buttons made of flexible fabric sensors have good wearing comfort. A sensor device with a simple structure is proposed. The sensor device can sense finger pressing actions within 12kPa. The micro control unit used by the controller is ESP32 which is used for sensor signal acquisition and data processing. Using the integrated Bluetooth chip inside the ESP32, the controller can connect to different terminal devices across platforms for wireless data transmission. The results show that the wireless music controller made in this article can be connected stably with both the Windows computer and the Android mobile phone. The accuracy of the sensor's recognition of finger pressing is 99.7%, shows that flexible fabric sensors have broad application prospects in the field of wearable devices.
Due to the impact of COVID-19, the previous centralized customer service has been changed to a home office model. Traditional distributed Call-Centers cannot solve the problem of multiple users sharing platform resources. This article analyzes its causes from two dimensions: market and technology. This is the research background of this article. The purpose is to study a Call-Center that not only meets the needs of remote work by agents, but also enables centralized data management. The distributed rental Call-Center platform born from this is a feasible solution to solve the problem of multi tenant shared call resources. The significance of the result lies not only in its progressiveness technology, but also in that the platform adapts to the development trend of the future Call-Center in outsourcing services, agents leasing and other aspects..
Aiming at the demand of real-time detection of site helmet wearing, an artificial intelligence application based on android embedded system is designed to promote the accurate management of smart site safety. The dataset is obtained by Internet crawlers and construction site videos. Samples are divided into training set, validation set and test set in the ratio of 8:1:2. The yolov5s is used as the objective detection algorithm, and the network to be modified. After training the model is quantified, and then the model calls the interface and the lightly quantified model to improve the inference speed of the model. The accuracy of the trained model inference is 92%, the mAp is 80.4%, and the detection speed is 38s. After quantization and deployment to android system, the accuracy is 89.2 and the mAp is 87.6%. The detection speed is 19.2ms, which maximizes the detection speed under the premise of ensuring accuracy. The design helps to detect the safety helmet worn by site personnel automatically and promote the safety management of personnel at smart sites.
In this paper, Baidu voice recognition method is adopted to study a sorted garbage bin.The main conclusions can be summarized as follows: (1) The experimental results show that the voice interactive sorting garbage bin can accurately and efficiently identify the garbage described by the users.(2) It solve the problem that the residents do not know how to sort the garbage. The voice interaction sorting garbage bin can also detect the volume information of the garbage in the bin and upload it to the city brain to optimize the scheduling of garbage recycling trucks. In terms of the future work, sorted garbage bin should be carried out to enhance for the study of sorting garbage bin can industrialization.
With the development of wireless communication, positioning and sensor technology, the acquisition of spatiotemporal trajectory data becomes more and more easy. Spatiotemporal trajectory data is composed of a series of trajectory points including location, time, speed, heading and other information, which contains rich spatiotemporal dynamic information of mobile objects. In this paper, we propose an improved algorithm for density peak clustering CFSFDP, which uses Gaussian kernel function and information entropy to optimize neighborhood search radius. This method can accurately mine routine behaviors from a large number of trajectory data.
This paper proposes a three-circle triangle positioning adjustment model for bearing-only passive locating in circular formation of UAV. Considering the realistic factor of transmitter position bias, the model stipulates the use of three signal transmitters and analyzes the relative position relationship of the transmitter and the target UAV to determine the trajectory circles. Using the triangle formed by the intersection points of the circles, the model utilizes the inner center to determine the most probable position of the target UAV. The UAV formation adjustment scheme is then designed based on the principle of maximizing accuracy and corrected priority. The model presented in in this paper has high positioning accuracy, wide applicability and strong generalization.
At present, there is still a lack of clear understanding of the underlying logic and development picture of the application of big data technology embedded in university party building. The prescriptive requirements and main trends of Party building in colleges and universities should be concentrated in the aspects of building intelligent innovation, strengthening the leadership of the Party, strengthening political construction, applying new technologies and building system. This paper establishes a rule-driven design for the metadata of Party building information in colleges and universities based on big data, and designs different rules for each requirement of the client and background server through classification. Specifically, there are interface display rules, storage rules, query rules, data item rules, etc. These rules are recorded in the corresponding rule files, which are analyzed by a unified rule parser when the system is running. Get the data characteristics of the corresponding function.
The development speed of information technology is remarkable. At first, information technology only intervened a lot in the field of industry. Now, with the improvement of people’s overall economic level and the development of Internet technology, all kinds of emerging technologies have been widely used in various industries, and many families have bought relevant facilities, greatly improving the quality of life. The author attempts to combine the latest cloud computing technology to explore its application in the Internet of things smart home system, and make theoretical contributions to the development of smart home and the promotion of information technology in China, so as to meet people’s new needs for life.
With the outbreak of the COVID-19, there have been increasing demands for the accurate and fast detection of the virus. The high-throughput liquid handling workstations have been widely used in the process of nucleic acid testing (NAT) because of their high automation and great pipetting abilities. The software for the workstations plays an indispensable role in the experiment design and the work management. However, most of the software for the workstations cannot meet the need of the openness and the security when designing the experimental procedure. As a result, the workstations can only assist the experienced operators in conducting the limited experiments. Aiming at solving these problems, the software for the high-throughput liquid handling workstations has been developed based on the principle of "check with design". During the workstation configuration and procedure edition, the properties of the consumables and the experimental steps are logically and quantitatively verified to help the operators edit the complete experiment successfully. The experiment in this research verified that the software can narrow the gap between operators with different professional background and improve their experiment design efficiency. When the infectious diseases break out, the relevant operators can be quickly mobilized to take part in the virus detection.
As an important content of rural governance, the modernization of rural public service governance is the need to complement the weaknesses of rural basic public service and improve the quality of public service. It is also an important basis for ensuring and improving people's livelihood and realizing rural revitalization. After clarifying the key governance objects and supply list of rural public services, the intelligent and digital construction of rural public service governance is well promoted, and the multi-subject collaborative governance of rural public service governance is realized. In this paper, an optimized Apriori algorithm is proposed for the massive data generated by the existing rural public service governance "one-stop" system. The optimized Apriori algorithm solves the problem that the efficiency of massive data mining in the system is not affected by the growth of massive data volume. The performance of the improved algorithm becomes more prominent when the support degree is less, and the advantages become more obvious.
The K-means algorithm is a common clustering algorithm for data mining, which has a slow iteration speed when dealing with large-scale data, and the initial cluster center selection also has a large impact on the clustering results. Therefore, this paper proposes an improved K-means algorithm based on the spark framework, and the performance of the algorithm is experimentally verified. The experimental results show that the improved K-means algorithm based on the Spark framework has good clustering effect and accuracy.
With the further development of globalization of business, informatization of language services and networking of enterprise contents, terminology management is an essential part of current translation projects. The role of terminology management is increasingly important for enterprises and independent translators, such as unifying and specifying terminology before translation, collecting, modifying and expanding terminology entries and quickly searching terminology during the translation process, and checking terminology consistency after translation. In this paper, we take SDL Trados Studio 2017, a computer-aided translation software, as an example to discuss the problems and solutions in constructing and applying terminology library.
In this paper, we present the LCCEN, a novel lightweight and crisp curve extraction network. By leveraging two state-of-the-art approaches, the LCCEN achieves comparable performance to heavy architectures, while requiring less than 4% of the parameters of existing methods. The proposed architecture delivers superior performance when compared to lightweight models, which typically contain fewer than one million parameters, by generating finer curve edge maps. We present a comprehensive analysis of the curves datasets by providing both qualitative and quantitative extraction results and compare the performance of LCCEN with state-of-the-art models. Moreover, LCCEN is designed without relying on pre-trained weights and has simple hyperparameter settings.
At present, the precision teaching supported by intelligent technology has some shortcomings in practice, such as single technology bias, lack of thinking of educational cognitive logic, and weakening of teachers' role status. One of the reasons lies in the lack of theoretical perspective on the academic logic of technology-enabled precision teaching. From the perspective of educational cognition, on the basis of clarifying the concept connotation of precision teaching, the research analyzes the existing practical difficulties in the process of its development, such as goal setting, diagnostic evaluation and activity development. Based on the function analysis of intelligent technology, the realization logic of intelligent technology-enabled precision teaching goal, evaluation and activity accurate development is discussed: With the help of group analysis, individual needs diagnosis, mapping relationship analysis to achieve accurate target setting; With the help of data acquisition, analysis mining and visualization technologies, the automatic implementation and intelligent analysis of evaluation are realized to facilitate the hierarchical and precise intervention of the application of evaluation results. With the help of resource recommendation and tool assistance, it can help the accurate design and practice of multiple learning activities, and provide support for the process intervention and real-time feedback of learning activities.
This paper mainly focuses on detecting the image data of sports football in the image collected by the calibrated camera, and then researches the location of the center of the circle in the camera image. Firstly, the basic principles and characteristics of several mainstream moving object detection methods are introduced: background difference method, frame difference method and optical flow method. By comparing the differences of these methods, the most suitable method for this system is selected for experimental analysis. Then, based on the detection of the moving object image, the football moving image is denoised, feature extraction and the football imaging center calibration, and the experimental results are analyzed.
