10%

Try : Insurtech, Application Development

Edtech(5)

Events(31)

Interviews(10)

Life@mantra(10)

Logistics(1)

Strategy(14)

Testing(7)

Android(43)

Backend(28)

Dev Ops(2)

Enterprise Solution(20)

Frontend(28)

iOS(39)

Javascript(13)

AI in Insurance(26)

Insurtech(57)

Product Innovation(34)

Solutions(13)

Augmented Reality(7)

Customer Journey(7)

Design(6)

User Experience(21)

Artificial Intelligence(95)

Bitcoin(7)

Blockchain(14)

Cognitive Computing(7)

Computer Vision(6)

Data Science(14)

FinTech(41)

Intelligent Automation(25)

Machine Learning(43)

Natural Language Processing(10)

E-health(2)

HealthTech(5)

mHealth(3)

Telehealth Care(1)

5 Insurance Front-Office Processes You Can Improve with AI

6 minutes, 5 seconds read

Amidst the growing footprint of Insurtech around the world, Insurance service models continue to evolve for both front and back-office processes. Currently, InsurTechs are using AI in three main areas: Customer Experience (58%), Product Innovation (43%), and Process Improvement (19%) — according to a McKinsey report. An organization’s ‘Front Office’ strategy will need to embody intelligent sales force automation, call-centre management, help-desk applications, product configuration and risk assessment tools. Insurance Carriers are restructuring these operations with an outward focus — aimed at improving interactions with their customers. 

While the Insurance back-office is focussed on streamlining in-house operations, the front office is responsible for driving customer experience, engagement and behaviour. However, most front-office operations deal with repetitive customer-facing jobs. Using Artificial Intelligence-based technologies such as RPA, tasks that require human mediation can now be handed over to automation technologies that imitate human interactions. Gartner estimates 20% of RPA will be cloud-based by 2022.

The real benefit of undergoing automation transformation is that both the front & back office can now be contextually linked in a smart manner — avoiding ‘working in isolation’ for extended periods. Customer-facing agents and reps can access information across the back-end more reliably and faster than before. Automating even routine tasks such as updating customer information, performing security checks, fetching product details or updating complaint forms — can reduce resolution times and the potential for manual errors.

This allows the front-office staff to focus on the most pressing matter — the relationship with the customer.

Customer servicing can now take place at incredible scale and complexity using chat, mobile and voice self-service tools. For example, speech recognition can capture what type of service to offer the customer (eg: update contact information, access policy details etc). These tools can also detect ‘anger’ or ‘frustration’ from the tone of voice and the information is passed to front-line reps who can quickly resolve an issue. As a result, remote diagnostics and self-service tools will see enhanced adoption over the coming years. The market for AI-enabled technologies in the claims process alone will be worth $72B by 2020.

5 key front-office operations that can be improved with AI

  1. Underwriting
    The most central function within the insurance value chain is to price risk. Using AI, the insurance underwriting process is now empowered with real-time insights derived from models analysis tons of customer-centric data.

    Using historical data, machine learning models can be trained to understand ‘known risks’ based on experience. For ‘unknown risks’, IoT sensors play a crucial role — by delivering a real-time picture of an ongoing operation. This allows for a second model to infer risk based on current data and the entire historical record of that specific process.

    Armed with in-depth knowledge about risk, insurers are moving from traditional risk pricing to a more proactive risk mitigation role. Through this new approach, carriers can set up real-time risk alerts, predict fraud and more accurately forecast ‘claims occurrence’ across the customer life cycle.

  2. Policy Administration
    A policy administration system is a backbone that manages all the policies within an insurance company. From the first point of interaction to fetching data from the back-office — most, if not all core operations run through this system. However, most insurance organizations still rely on legacy systems that require tremendous workaround using manual efforts.

    According to a study by Celent, nearly 45% of Insurance CIOs identified disconnected and duplicative legacy systems as a key inhibitor to digital transformation.

    Today’s challenging market dynamics and competitive pricing pressures are changing this approach. There are several areas worth investing in for carriers such as image & voice recognition to capture and authenticate customer information at the initial contact stage to intelligent entity extraction tools for understanding even handwritten text from a physical document.

    Automation enhancements help drive policyholder retention by improving connectivity to the back-end and delivering the most optimal outcomes for front-office workflows.

  3. Claims Management

    Claims are the most widely scrutinized function within the insurance value chain. Most claims servicing is performed by human agents over the phone. With speech recognition, these conversations can be automatically transcribed/ translated in real-time. This frees up more agent time to handle greater issues while leaving automation enabled self-service to handle the most basic customer queries.

    Claims assessment or loss estimation itself can be performed remotely using image recognition tools linked to algorithms that can calculate the payout for the policyholder.

    Without the need for human intervention, straight-through processing can be dramatically improved by reducing processing time — allowing human agents to react faster to policyholders demands.

    Also, read – How AI can settle claims in 5 minutes!

  4. Marketing & Sales Distribution
    According to Salesforce, only 36% of the average salespersons’ week is spent selling. Human sales reps typically spend a large portion of their time nurturing unqualified leads. With sales funnel maximizers, like LCA, reps can get quick access to leads that have been scored, prioritised and allocated for the right agent to optimize conversions.

    Distribution and sales chains are moving to a completely digital and affinity-based ecosystem. Chatbots and virtual agents can, therefore, play a critical role in increasing cross-sell and up-sell opportunities. These AI-enabled tools are fitted with Natural Language Processing (NLP) capabilities to contextually interpret the interaction with the customer.

    AI also leverages predictive analytics to produce behavioural insights when pitching the customer — allowing the agent to ask the right questions, address unmet needs and resolve anticipated near-term challenges.

  5. Product Personalization
    Using Machine Learning algorithms to precisely price risk, allows Carriers to understand the complexities involved in new product development — especially measuring the ‘unknown risks’ involved in creating new product lines.

    Data (both historical and IoT derived) coupled with predictive analytics can offer more personalised guidance to insurance buying. InsurTechs are poising themselves strategically in this area, ahead of the large carriers, to attract a new and younger customer base. Companies like MetroMile, Trov and Lemonade have been able to create unique offerings with AI-derived insights fine-tuned to the individual, while also charging much lower premiums than the market.

    New customers are able to buy convenient, sachet-type, even pay-as-you-use modelled insurance products for protecting their assets (mobile, laptop, home appliances, short travel, vacations etc). This has brought about an appetite for on-demand insurance where insurance can be bought, queries can be resolved and claims can be processed, all within a few minutes.

Other Customer-Facing Areas improved by AI

1. Proactive Front-Office Processes 
2. Precise Risk Mitigation/Active loss prevention
3. Chatbots and Robo-advisors 
4. Real-time Underwriting 
5. Accurate Claims Processing 
6. Direct Marketing & Cu0stomer Retention
 7. Bespoke Insurance Advice
 8. Understanding User’s Emotions 

Forrester predicts the impact of intelligent automation — through evidence in ‘the service desk’. They claim: automation will eliminate 20% of all service desk interactions, by the end of 2019. Enabling human workers with digital assistants in the insurance front-office has scope for very high disruption. Human agents are prone to making repeat errors that automation equipped with AI can fix easily — especially in routine and repetitive tasks.

Carriers, now have the opportunity to boost their market position by improving agent productivity, reducing operational inefficiencies like reprocessing, producing errorless transactions for customers and thereby creating an uninterrupted service chain.
Mantra Labs solves the most challenging front & back-office operations plaguing the Insurance value chain. To know more about our work in this space, reach out to us on hello@mantralabsglobal.com.

Cancel

Knowledge thats worth delivered in your inbox

Tabular Data Extraction from Invoice Documents

5 minutes, 12 seconds read

The task of extracting information from tables is a long-running problem statement in the world of machine learning and image processing. Although the latest accomplishments in the field of deep learning have seen a lot of success, tabular data extraction still remains a challenge due to the vast amount of ways in which tables are represented both visually and structurally. Below are some of the examples: 

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Invoice Documents

Many companies process their bills in the form of invoices which contain tables that hold information about the items along with their prices and quantities. This information is generally required to be stored in databases while these invoices get processed.

Traditionally, this information is required to be hand filled into a database software however, this approach has some drawbacks:

1. The whole process is time consuming.

2. Certain errors might get induced during the data entry process.

3. Extra cost of manual data entry.

 An invoice automation system can be deployed to address these shortcomings. The idea is to upload the invoice document and the system will read and generate the tabular information in the digital format making the whole process faster and more cost-effective for companies.

Fig. 6

Fig. 6 shows a sample invoice that contains some regular invoice details such as Invoice No, Invoice Date, Company details, and two tables holding transaction information. Now, our goal is to extract the information present in the two tables.

Tabular Information

The problem of extracting tables from invoices can be condensed into 2 main subtasks.

1. Table Detection

2. Tabular Structure Extraction.

 What is Table Detection?

 Table Detection is the process of identifying and locating tables that are present in a document, usually an image. There are multiple ways to detect tables in an image. Some of the approaches make use of image processing toolkits like OpenCV while some of the other approaches use statistical models on features extracted from the documents such as Text Position and Text Characteristics. Recently more deep learning approaches have been used to detect tables using trained neural networks similar to the ones used in Object Detection.

What is Table Structure Extraction?

Table Structure Extraction is the process of extracting the tabular information once the boundaries of the table are detected through Table Detection. The information within the rows and columns is then extracted and transferred to the desired format, usually CSV or Excel file.

Table Detection using Faster RCNN

Faster RCNN is a neural network model that comes from the RCNN family. It is the successor of Fast RCNN created by Ross Girshick in 2015. The name Faster RCNN is to signify an improvement over the previous model both in terms of training speed and detection speed. 

To read more about the model framework, one can access the paper Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.

 There are many other object detection model architectures that are available for use today. Each model comes with certain advantages and disadvantages in terms of prediction accuracy, model parameter size, inference speed, etc.

For the task of detecting tables in invoice documents, we will select the Faster RCNN model with FPN(Feature Pyramid Network) as a feature extraction network. The model is pre-trained on the ImageNet corpus using ResNET 101 architecture. The ImageNet corpus is a public dataset that consists of more than 20,000 image categories of everyday objects.  We will therefore make use of a Pytorch framework to train and test the model.

The above mentioned model gives us a fast inference time and a high Mean Average Precision. It is preferred for cases where a quick real time detection is desired.

First, the model is to be trained using public datasets for Table Detection such as Marmot and UNLV datasets. Next, we further fine-tune the model with our custom labeled dataset. For the purpose of labeling, we will follow the COCO annotation format.

Once trained, the model displayed an accuracy close to 86% on our custom dataset. There are certain scenarios where the model fails to locate the tables such as cases containing watermarks and/or overlapping texts. Tables without borders are also missed in a few instances. However, the model has shown its ability to learn from examples and detect tables in multiple different invoice documents. 

Fig. 7

After running inference on the sample invoice from Fig 6, we can see two table boundaries being detected by the model in Fig 7. The first table gets detected with 100% accuracy and the second table is detected with 99% accuracy.

Table Structure Extraction

Once the boundaries of the table are detected by the model, an OCR (Optical Character Reader) mechanism is used to extract the text within the boundaries. The text is then processed using the information that is part of a unique table.

We were able to extract the correct structure of the table, including its headers and line items using logics derived from the invoices. The difficulty of this process depends on the type of invoice format at hand.

There are multiple challenges that one may encounter while building an algorithm to extract structure. Some of them are:

  1. The span of some table columns may overlap making it difficult to determine the boundaries between columns.
  2. The fonts and sizes present within tables may vary from one table to another. The algorithm should be able to accomodate for this variation.
  3. The tables might get split into two pages and detecting the continuation of a table might be challenging.

Certain deep learning approaches have also been published recently to determine the structure of a table. However, training them on custom datasets still remains a challenge. 

Fig 8

The final result is then stored in a CSV file and can be edited or stored according to one’s convenience as shown in Fig 8 which displays the first table information.

Conclusion

The deep learning approach to extracting information from structured documents is a step in the right direction. With high accuracy and low running time, the systems can only learn to perform better with more data. The recent and upcoming advancements in computer vision approaches have made processes such as invoice automation significantly accessible and robust.

About the author:

Prateek Sethi is a Data Scientist working at Mantra Labs. His work involves leveraging Artificial Intelligence to create data-driven solutions. Apart from his work he takes a keen interest in football and exploring the outdoors.

Further Reading:

Cancel

Knowledge thats worth delivered in your inbox

Loading More Posts ...
Go Top
bot

May i help you?

bot shadow

Our Website is
Best Experienced on
Chrome & Safari

safari icon