Periodic Reporting for period 1 - SentiSquareCX (Qualitative Jump in Customer Experience: Omnichannel Impact of Distributional Semantics)
Reporting period: 2019-06-01 to 2019-11-30
"In a world where customer experience (CX) is becoming the #1 differentiator, the ability to monitor and manage inbound customer communications is crucial to commercial success. That is requires the deployment of Natural Language Processing (NLP) to understand, classify, direct, monitor, and respond to customer queries. SentiSquare helps contact centres deal with the increasing traffic by boosting and automating processes in channels like voice (transcribed), chat, email, tickets, and more. SentiSquare uses machine learning and distributional semantics to provide analysis of any and all customer-generated text, no matter the language. The objective of SentiSquare is to scale up and deliver the product to new markets and industries."
Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far
During the duration of this project, we made considerable progress towards scaling up, and several key achievements have been accomplished. To achieve good market fit, we have validated the competitiveness of our products with several large clients in telco, finance, and utilities. We have also deployed a brand new solution assessing the quality of agent communication with clients.We have also developed our ability to classify phone call transcripts, completed a chatbot support pilot project, and upgraded our Analytics user interface. On top of that, we have formulated a strategy to move towards a Software-as-a-Service model which will allow us to expand faster through speeding up the sales and implementation process.
Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)
To extend our technological edge, we are developing cross-lingual learning, which removes the need to build models for each language, as well as Fuzzy Classification, which builds an inferred semantic model based on little input from the client, allowing our products to be deployed immediately, without the client’s data. The combination of these new technologies has the potential to deliver the benefits of Natural Language Processing languages in which the use of the technology has been hitherto limited due the shortage data. In the European context, that means that smaller economies will be able to benefit economically at the same rate as the larger ones with access to state-of-the-art NLP for large languages. Through the streamlining of customer care processes, SentiSquare’s technology can also contribute to the increase of productivity as well as profitability of companies.