Con esta herramienta te facilitamos un acceso a todas las ofertas y demandas de tecnología europeas y a búsquedas de socios para participar en propuestas europeas de I+D publicadas en la red Enterprise Europe Network, pudiendo filtrar los resultados para facilitar las búsquedas más acordes con tus necesidades.

¿Quieres recibir estos listados de oportunidades de colaboración en tu correo de forma periódica y personalizada? Date de alta en nuestro Boletín

Para optimizar los resultados de la búsqueda, se recomienda utilizar términos en inglés.

Solución de análisis multilingüe de opiniones del cliente

Resumen

Tipo:
Oferta Tecnológica
Referencia:
TOES20180613001
Publicado:
22/06/2018
Caducidad:
23/06/2019
Resumen:
Una empresa española de base tecnológica con una sucursal en EE. UU. (el 70% de su facturación procede de Silicon Valley) ha desarrollado una plataforma de análisis multilingüe de opiniones del cliente. Esta plataforma recoge y analiza textos desestructurados de redes sociales, reseñas en internet, registros de centros de llamadas y cualquier tipo de texto para transformarlo en opiniones valiosas y tomar decisiones comerciales. La empresa utiliza los avances más recientes en procesamiento del lenguaje natural (PLN) y capacidad computacional. Actualmente existen otras soluciones en el mercado pero son limitadas en cuanto al número de idiomas que soportan y además no combinan estadística con conocimiento del lenguaje (uso coloquial). La empresa busca socios con el fin de establecer acuerdos de licencia.

Details

Tittle:
Multilingual customer review analytics solution
Summary:
A Spanish tech-based company, with branch in USA (70% from its turnover from Silicon Valley), has developed a multilingual text analytics customer experience platform. It collects and analyses all the unstructured texts from social media, online reviews, call centers logs, and any kind of text in order to transform it into valuable insights to take business decisions. The company is looking for license agreements.
Description:
Knowing customers opinion is a crucial information for the development of any company. The problem is that, most of the people show their opinions about products and services on the internet (blogs, forums, etc.). It is a big amount of unstructured information that is difficult to collect and to analyze butessential to take the right decisions.

In order to address this problem, the Spanish company has developed a tool that collects a vast amount of texts (from social media, online reviews, etc.) and transforms it into a few insights that are relevant for business decisions. The company uses the most recent advances in NLP (Natural Language Processing) and computational power, learning any given brand´s global reputation is almost totally automated.

Currently, there are other solutions in the market that provide customer review analytics but are limited regarding the number of languages supported and do not combine statistics with language knowledge (coloquial use of language). On the other hand, this sloution supports more than 20 languages and combines statistics with language knowledge to provide a more accurate analysis.

The company was born in 2008 and now is also established in the USA (70% of the company turnover).

The solution presents the following features:
-Text structuration: extracts precisely what customers are feeling about particular topics - and even how strongly they feel - with over 70% (out-of-the-box) and 90% (after fine tuning) accuracy across +20 languages and variants.
-Cost effective: 1/3 of the manual cost.
-Priorities definition: It helps to create new categories on the go so users can focus on what is key for them, like specific stores, staff, pricing strategy, locations, promotions, products or features.
-Topic-based analysis (not document or sentence -based). Picture attached.
-Multilingual: +20 languages and variants (Turkish, Russian,...).
-Categorization: 420 categories (sales, aftersales, brand...).
-Bottom-up: Extracts the more frequent topics and concepts.
-Spelling: Context dependent and context independent.
-Removal of proper names, brands and offensive words, helping to comply with GDPR (General Data Protection Regulation) legislation.
-Language detection: Language identification in documents that may contain text from more than one language or variant.
-Flexible and customizable visualization.

The solution is offered in four different packs depending on needs:
1) General purpose pack (without categorization):
- 8 languages.
- Topic-based sentiment analysis: To identify opinions at the topic level, not the document or sentence.
- Entity extraction:Comply with GDPR (General Data Protection Regulation).
- Topic/phrase extraction: Identifies the main issues in the reviews.

2) Hospitality sector pack:
- 8 languages.
- Topic-based sentiment analysis: To identify opinions at the topic level, not at the document or sentence.
- Categorization: Fine-grained approach allows for the handling of customized categories.
- Entity extraction: Comply with GDPR (General Data Protection Regulation).
- Topic/phrase extraction: Identifies the main issues in the reviews.

3) Automotive sector pack:
- 20 languages.
- Topic-based sentiment analysis: To identify opinions at the topic level, not at the document or sentence.
- Categorization: Fine-grained approach allows for the handling of customized categories.
- Topic/phrase extraction: Identifies the main issues in the reviews.

4)Consulting pack for producing different sector and languages:
- It allows to customize the solution depending on needs and challenges.

The SaaS version of the platform is aimed at customer experience officers, marketing executives, and VoC (Voice of the Customer) champions, specially, those working for the retail sector.

The company is looking for international ICT partners willing to cooperate under license agreement in order to market this technology within a defined area.
Advantages and Innovations:
-There are similar tools in the market but this is the only one that works with more than 20 languages and combines statistics with language knowledge.

-It takes the guesswork out of market research and customer insight, revealing trends that appear across the entire data collection.

-No programming skills required: Excel-based and designed for business users.

-Fully automated process that provides big savings if compared to manual analysis (automatic analysis is around 1/3 of the cost).
Stage of Development:
Already on the market
IPs:
Secret Know-how

Partner sought

Type and Role of Partner Sought:
The company is looking for international ICT integrators with good knowledge of the client oriented business sectors (retail, hospitality, automotive,...), market research companies and CX (Customer Experience) platform providers willing to market this technology under license agreement.

Client

Type and Size of Client:
Industry SME 11-49
Already Engaged in Trans-National Cooperation:
Si
Languages Spoken:
English
Spanish

Keywords

Technology Keywords:
01003006 Computer Software
01003015 Gestión de conocimiento, gestión de procesos
01005004 Human Language Technologies
01003013 Information Technology/Informatics