What are the characteristics of the application of artificial intelligence in daily life (what are t

1. Artificial intelligence software chat GPT

There are nearly a hundred answers now, but no one mentions an artificial intelligence application that almost every Chinese person comes into contact with every day - content auditing With the introduction of deep learning, content review technology has undergone rapid development, which is faster than facial recognition, navigation, content recommendation, and other technologies that can be seen in daily life, with richer applications and more diverse forms.

2. Artificial Intelligence Movies

At present, it is far beyond the imagination of ordinary people. Let's first talk about the general form of content review, which is still in the imagination of ordinary people. The technology is relatively simple and often perceived, including but not limited to: keyword review [2], which means that "above" and the online community each establish and maintain a keyword blacklist, and then use keyword hit to conduct the review.

3. List of artificial intelligence stock leaders

Of course, keyword hits may lead to unexpected name combinations being mistakenly hit (typically, AB is a keyword, but CA and BD are each a phrase, which can cause incorrect hits when the CABD sequence appears). Therefore, some online communities have introduced intelligent word segmentation techniques for keyword review to greatly avoid such incorrect hits

4. Artificial intelligence AI

[3] [4], reduce users' perception of being audited 2 Topic review [5] Topic review is generally conducted through topic indexing and hashtags. It is applied to top topics and usually involves manual review. Without review, any content including corresponding hashtags will appear on other users' timelines, but after the introduction of review, targeted content will be conducted.

5. Artificial intelligence computing power

Exposure control, for content that does not comply with the direction of caliber control, creates a situation where the author is visible but the followers are not, or reduces content distribution/exposure to eliminate or reduce its impact and reduce the perception of the audited users After the event, manual intervention review, deletion, or reduction of exposure/sharing is carried out based on the activation condition of a sudden increase in the exposure of a certain content in sensitive populations.

6. Artificial Intelligence English

4. The above are relatively simple forms. After deep learning and computing power unlocking, we are experiencing some new forms, including but not limited to 1. Social graph network user profiles have billions of vertices and over trillion edges based on the chat interactions, family relationships, contact lists, buying and selling relationships, geographical relationships, business relationships, and other components between users

7. Artificial Intelligence GPT

Social graph network [6], in which a very small proportion of sensitive users can be calibrated to accurately depict a population of billions, including their personality, political orientation, speech style, influence, etc. Based on this, a small number of sensitive groups can be dynamically designated as the focus of human intervention, and even their content on all online platforms such as WeChat, Weibo, and phone can be reviewed indiscriminately.

8. Artificial Intelligence and Intelligent Manufacturing

2. Understanding and Intention Determination of Multimodal Information Content [7] The adjacency of text, images, videos, voice, and other content on the timeline means that they have content relevance between them. Multimodal information content understanding can more accurately extract content information from timeline content such as chat, social media, ideas, Weibo, etc., determine its sensitivity, and analyze the fundamental intentions of users participating in timeline discussions

9. Ranking of the Best Schools in Artificial Intelligence

[8] [9], thereby locking sensitive users in near real-time without human intervention and conducting targeted supervision 3 Traditionally, information flow distribution intervention understands that review is a passive action that needs to be made after users publish content, but information flow distribution intervention can expose more and implicitly to users some controlled content that does not cause their aversion without users' awareness, thus changing users' perception of society and their political orientation in the long run.

10. Artificial Intelligence Big Model

[10] It has a positive and stable impact on user speech [11], corrects users' misconceptions and incorrect memories, and can even achieve the ultimate goal in the medium to long term, effectively reducing audit computing power and manual audit pressure, maintaining social civilization and stability, and promoting social harmony Attention shift.

When there are news hotspots that cannot be reconciled under mainstream values and ideological systems, pushing other controversial hotspots to users effectively distracts their attention [12] [13], reduces user perception disturbance, and thus reduces the possibility of extreme speech output, maintaining social stability 5 With the assistance of deep learning technology, auditing no longer exists as fixed keywords, but in a dynamic form.

这种情况下，用户无法对审核规则进行准确感知，提高了审核的不可知性，降低了用户与审核对抗的意图审核技术日新月异，限制它的不是算力和技术，而是我们的想象力参考^Xu, B., & Albert, E. (2014). Media censorship in China. Council on Foreign Relations, 25, 243.

^Xu, X., Mao, Z. M., & Halderman, J. A. (2011, March). Internet censorship in China: Where does the filtering occur?. In International Conference on Passive and Active Network Measurement (pp. 133-142). Springer, Berlin, Heidelberg.

^Huang, H., Wang, X., & Wang, H. (2020). NER‐RAKE: An improved rapid automatic keyword extraction method for scientific literatures based on named entity recognition. Proceedings of the Association for Information Science and Technology, 57(1), e374.

^Song, Y., Kim, E., Lee, G. G., & Yi, B. K. (2004). POSBIOTM-NER in the Shared Task of BioNLP/NLPBA2004. In Proceedings of the International Joint Workshop on Natural Language Processing in Biomedicine and its Applications (NLPBA/BioNLP) (pp. 103-106).

^Liao, S. (2019). “# IAmGay# What About You?”: Storytelling, Discursive Politics, and the Affective Dimension of Social Media Activism against Censorship in China. International Journal of Communication, 13, 21.

^Chaudhary, A., Mittal, H., & Arora, A. (2019, February). Anomaly detection using graph neural networks. In 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon) (pp. 346-350). IEEE.

^Dey, N., Ashour, A. S., & Nguyen, G. N. (2020). Recent advancement in multimedia content using deep learning.

^Froehlich, D. E., Rehm, M., & Rienties, B. C. (2020). Mixed methods social network analysis. Mixed methods social network analysis: Theories and methodologies in learning and education, 1-10.

^Hu, M., Peng, J., Zhang, W., Hu, J., Qi, L., & Zhang, H. (2021). An intent recognition model supporting the spoken expression mixed with Chinese and English. Journal of Intelligent & Fuzzy Systems, 40(5), 10261-10272.

^Sparr, M. (2022). Explicit User Manipulation in Reinforcement Learning Based Recommender Systems. arXiv preprint arXiv:2203.10629.

^Papakyriakopoulos, O., Serrano, J. C. M., & Hegelich, S. (2020). Political communication on social media: A tale of hyperactive users and bias in recommender systems. Online Social Networks and Media, 15, 100058.

^Paasonen, S. (2016). Fickle focus: Distraction, affect and the production of value in social media. First Monday.

^Xie, J. Q., Rost, D. H., Wang, F. X., Wang, J. L., & Monk, R. L. (2021). The association between excessive social media use and distraction: An eye movement tracking study. Information & Management, 58(2), 103415.