"As the world moves towards automation, a growing market across multiple sectors is 'automated advice', a key demand for which, as highlighted by HM Treasury and the FCA, is in the financial advice sector resulting from the complexity and ever-changing nature of regulations/legislation. An example is the difficulty in providing a holistic automated advice solution for financial advice due the requirement for human interpretation of the rules to a specific case. In order for an automated advice model to be successful, the algorithm underpinning it must be completely accurate in the application of expert knowledge, be able to address every possible scenario, and also provide an audit trail of how the ultimate recommendation was determined. Consequentially little innovation has occurred in automated advice provision in more complex tax areas, such as innovation targeted tax incentives, due to the difficult interpretation of guidance to widely varying scenarios, which cannot be modelled via standard flowchart approach. This has led to many SME's failing to benefit due to a lack of knowledge, fear of engaging directly with HMRC, or high cost barriers to explore potential schemes. DittoAI have developed a unique knowledge acquisition methodology, free from interviewer dependency that can produce a computable output that exhaustively documents an individual's decision-making & encodes their entire expertise on any subject. DittoAI can guarantee a complete, correct and consistent output for any subject matter, along with a complete audit trail to reveal all decision points, by digitally capturing the knowledge from any source and delivering precise expertise to the end-user via an artificial intelligence driven interface. **Impacts:** * The underpinning DittoAI technology is a hugely enabling technology needed by AI systems whenever they require detailed and traceable expert input to work, this project will allow further versions of the DittoAI technology to be exploited in other markets. * DittoAI offers ability for any company to access and codify the expertise of their few/ busy experts, and deliver it through any of their staff (or even in an automated fashion) increasing their service offering, delivery capacity, and removing the risks faced by losing key personnel from death/resignation, etc * Offering a reliable and consistent tax advice to ensure consumers are able to accurately complete their tax bills. Currently an estimated £10bil of potential R&D Tax Credits go unclaimed each year, due to the complexity of the guidance and incorrect/inconsistent advice offered by tax consultants."

Knowledge Base Systems (KBS) are vital for knowledge exchange & learning, proven to reduce costs, alleviate workforce shortfalls & stimulate innovation globally. The key limitation of KBS is Knowledge Acquisition (KA) - extracting expert knowledge in a computable format. Accuracy of KA is highly dependent on interviewer skill & expert ability to articulate knowledge, & projects are likely to fail at this stage causing a “Knowledge Acquisition Bottleneck”. Empiricom has developed, patented & commercialised a novel KA methodology, SOLAR Acquire (SA): free from interviewer dependency, it assists articulation via a structured interview process using Boolean (closed) questions to produce a computable output that exhaustively documents an individual’s decision-making & encodes their entire expertise on any subject. Benefits of SA are recognised by global orgs, however barriers prevent wider mkt uptake: lacks scalability, labour intensive requiring ~20 hrs of face-to-face interviews with experts at a high cost (£1k/day). Based on existing customer demand & extensive mkt research, Empiricom have identified a clear business opportunity in the development of a Natural Language Processing KA software assistant (NLP KAA), to enable Empiricom’s methodology to be automated thus removing barriers preventing wider exploitation in B2B apps where a significant mkt opportunity exists (global mkt worth $6.78bn in 2014). However, critical functionality of the KA for B2B apps requires further exploration, to investigate the feasibility of a NLP KAA with: - Full automation of the SA method such that, via an interaction with a machine, an expert in a complex subject area could successfully have “n-dimensional” knowledge captured from them - Natural language communication incl. correct translation of pronouns & grammaticallycorrect sentential reformulations. The project will result in PoC algorithms for the NLP KAA. Further prototyping will follow with expected mkt intro 2017

Empiricom has developed & commercialised a patented knowledge acquisition interview tool, SOLAR Acquire, which provides a structured interview technique & exhaustively documents an individual’s entire expertise on any subject. There is a clear opportunity for improving the accessibility of knowledge based systems (KBS) beyond B2B sectors, building on SOLAR Acquire, Empiricom seek to develop & validate components of an automated digital solution capable of extracting knowledge without human interviewers. This innovative solution will be freely available online (via Should-i.com social community platform) giving users the power to: create & distribute expert knowledge applets (small apps within an app), interact with applets to answer questions & participate in social networking. Working with subcontractors within Manchester’s creative & digital cluster, the project seeks to develop an assistant software program to capture knowledge, an inference engine, a community portal, a mobile platform & e-learning materials. All foreground IP generated around the assistant & inference engine will be owned & commercialised (mkt entry 2015) by Empiricom.