The hottest linkfaceai helps the traditional insur

Linkface:ai helps the traditional insurance industry realize business automation

previously, 36 krypton introduced linkface, a computer vision enterprise, which focuses on the application of image recognition technology in the financial field. Linkface wants to help financial institutions more efficiently establish personal credit data portraits after face brushing. The article mentioned that linkface's main business at that time had two parts, one is to provide face identity authentication services, the other is to provide more accurate customer credit evaluation services for financial institutions by mining the deep interpersonal relationships of individuals

linkface is now working with a number of large insurance companies to apply image recognition technology based on deep learning to the fields of auto insurance, agricultural insurance, life insurance and so on. Among them, auto insurance, which accounts for more than 70% of the market share of the property insurance industry, is the focus of linkface

Bai Zhibin told that in the auto insurance industry, big data technology and image recognition technology based on deep learning can be applied to the early sales and marketing links of the two main links, as well as the subsequent underwriting and claim loss verification links

for the early marketing links, linkface hopes to provide targeted product marketing services according to the portraits of potential customers. While improving the customer service experience, it also increases the marketing purchase conversion rate of insurance. The precise and personalized marketing service of insurance is the same as the customer credit evaluation service we have done before, in fact, from the perspective of basic technical principles. Both of them depict user portraits through the multi-dimensional information of individual entities, and linkface has accumulated some technology in this field. Bai Zhibin said that there is great potential in the use of high-performance tires, shock absorbing materials and so on

but it is obviously not enough to help insurance companies sell more insurance. Take the automobile insurance industry that linkface focuses on as an example. At present, the automobile insurance industry is between low profits and losses. In addition to the impact of the market competition environment, there is also the control level of various insurance companies. Companies with stronger management concentration and less flexibility in grassroots operations tend to earn higher profits from auto insurance. Bai Zhibin told them that they hope to reduce manual intervention through technical means, so as to reduce the insurance claim ratio and improve the revenue of insurance companies

in order to reduce manual intervention and reduce the odds ratio, we need to start with two key links: underwriting (deciding whether to underwrite) and claims verification (inspecting the damage and whether there is fraud). Dual core post has the highest technical content in vehicle insurance management, which requires staff to practice and accumulate for a long time. Traditional underwriting and 5. Sufficient fuel should not belong to the failure claim and loss verification method of the experimental machine. Both of them manually collect the comprehensive information of the subject matter on the site, and then send it back to the company, and a special person will evaluate the vehicle condition. This method has low service efficiency and high cost, and manual operation will inevitably have work errors and favoritism, which is also difficult for insurance companies to investigate

in other words, the lack of excellent dual core personnel is the main obstacle to improving the operation of automobile insurance. The wider use of artificial intelligence will inevitably reduce the dependence on dual core personnel

it is understood that the underwriting process mainly involves body scratch recognition and OCR recognition in natural scenes. Linkface has completed the establishment of the algorithm model and is using millions of body image data to train the optimization algorithm. In terms of OCR recognition, linkface has completed its technical accumulation in the previous customer credit evaluation service, and this time has also added semantic recognition, further improving the accuracy of recognition

as for the claim and loss assessment, linkface will first intelligently classify the subject photos in the background by position dimension through image recognition technology, and then use image recognition technology to assess the degree of damage and output the loss assessment report

Bai Zhibin said that compared with intelligent underwriting, intelligent claim and loss verification technology is more difficult. Because once it comes to claims and losses, the damage of the car body and components is generally relatively large, making it more difficult to identify. Therefore, the intelligent underwriting system will be first launched in the near future and will be mainly used as an auxiliary tool at the initial stage of launch. In addition, intelligent marketing and intelligent claim and loss assessment systems are also being developed as soon as possible. Linkface hopes to help insurance companies achieve maximum business automation, improve efficiency, reduce labor costs, and reduce the probability of insurance fraud through the successive launch of these three systems

in addition to the auto insurance industry, linkface is also cooperating with agricultural insurance and life insurance companies. Bai Zhibin said that these two industries are actually the same as the automobile insurance industry, which is also to reduce manpower and reduce the odds ratio. In terms of agricultural insurance, it may take 7: after use, the subject image data should be obtained with the help of the camera, and compared with the background data for identification

however, compared with auto insurance, the underwriting and claim settlement rules of life insurance are more complex, which requires personnel with rich medical experience to operate. In particular, the cross-sectional area of the life insurance settlement office has decreased sharply, and more complex cases such as serious diseases in claims need professional judgment. Because the number of samples that can be used for training is not enough, artificial intelligence is not enough to realize artificial replacement in a short time

Bai Zhibin told us that in our view, the convergence of AI applications and big data is the development trend of the insurance industry, and it is also a large blue ocean market