Conditional random fields (CRF) are widely used for sequence labeling such as Named Entity Recognition (NER) problems. Most CRFs, in Natural Language Processing (NLP) tasks, model the dependencies between predicted labels without any consideration for the syntactic specificity of the document. Unfortunately, these approaches are not flexible enough to consider grammatically rich documents like patents. Additionally, the position and the grammatical class of the words may influence the text’s understanding. Therefore, in this paper, we introduce SynCRF which considers grammatical information to compute pairwise potentials. Syn-CRF is applied to TRIZ (Theory of Inventive Problem Solving), which offers a comprehensive set of tools to analyze and solve problems. TRIZ aims to provide users with inventive solutions given technical contradiction parameters. SynCRF is applied to mine these parameters from patent documents. Experiments on a labeled real-world dataset of patents show that SynC RF outperforms state-of-the-art and baseline approaches.

Over the past decades, companies have continually sought out approaches that help them reduce the innovation cycle time due to its importance in their success. Among these approaches are TRIZ-based systematic inventive design processes, such as Inventive Design Methodology (IDM). Nevertheless, the application of these methods in the initial analysis , which requires an exhaustive gathering of information at the start of the innovation project without considering its impact on the final solution, decreases the agility of these types of processes. Consequently, a lean-based method called Inverse Problem Graph (IPG) has been proposed to formulate problems in the initial analysis phase of the inventive design process. However, the manual gathering of essential information to create a network of problems in the IPG method requires time and effort, which impresses the method’s capability. This paper integrates an automatic information retrieval approach, using Doc2vec and Cosine Similarity as Artificial Intelligence methods and Scientific Papers, into the IPG process. The integration helps to introduce a new method for the initial analysis phase of inventive design, helping to resolve part of its drawback in collecting essential knowledge from Scientific Data. The capability of the proposal is then tested through an application using a Lattice Structure case study .

In recent years, Deep Learning methods have become very popular in Natural Language Processing (NLP), especially transformer-based architecture. NLP domain requires a high volume of annotated data to work. Unfortunately, obtaining high-quality and voluminous labeled data is expensive and time-consuming. One promising method which has singled out for its performance in the context of data deficiency is semi-supervised learning with Generative Adversarial Networks (GAN). In this paper, we propose a new approach called PaGAN which is a combination of a document classifier and a sentence-level classifier inside a GAN for patent documents understanding. The idea is to mine the patent’s motivating problem (aka contradiction in TRIZ domain) which is fundamentally important to understand the underlying invention and its originality. PaGAN is applied and evaluated on a real-world dataset. Experiments show outperforming results of PaGAN comparatively to baseline approaches.

Su-Field analysis, as one of TRIZ analytical tool for solving inventive problems, can be used to improve the performance of the technical system effectively. Generally, choosing an appropriate inventive standard is critical to solving inventive problems efficiently and accurately. However, these standards are summarized and categorized based on the enormous amount of patents in different domains, and they are built in the high level of abstraction, independently of the specific application field, making their use require much more technical knowledge than other TRIZ tools. In order to facilitate the use of inventive standards, especially for capturing the uncertainty or imprecision depicted in the standards, a rule-based heuristic methodology is proposed in this paper. Firstly, Su-Field analysis ontology and fuzzy analysis ontology are built to represent the precise and fuzzy knowledge in the process of solving inventive problems respectively. Then, SWRL (Semantic Web Rule Language) inference and fuzzy inference are performed for generating heuristic concept solution. Finally, a prototype is developed and the resolution of the case of the “Auguste Piccard’s Stratostat” in prototype is elaborated in detail.

TRIZ theory is considered as a kind of innovative theory, which mainly functions in solving contradiction and aims to improve engineers creativity. It may reason on different support such as web and scientific documents. Most of the current knowledge, especially in industrial domains lies in patents. They are, nevertheless, an underused resource because of their complexity, their length and the need of domain knowledge to make use of patent. In this paper, we introduce an innovative application of deep learning to mine knowledge from patents. We show that our summarization based approach, called SummaTRIZ1, enables the extraction of these contradictions essential for TRIZ problem solving engine. A BERT based summarization approach is introduced to retain contradiction sentences. Our approach is experimentally evaluated on a real data set showing the performance of SummaTRIZ.

Nowadays, innovation represents one of the most crucial factors driving the success of companies. The Theory of Inventive Problem Solving (also known as TRIZ) is a well-established method to facilitate systematic inventive design. Although, TRIZ allows solving inventive problems through a panoply of knowledge sources, it may make inventive problem solving a time-consuming, experience demanding process and lead to waste of resources of the companies. To avoid the use of these tools and to help new users in solving their inventive problems without completely mastering TRIZ, we propose in this paper an approach based on the use of the Case-based reasoning (CBR) in order to capitalize experience. CBR is a knowledge paradigm that solves a new problem by finding the old similar cases and reusing them. The retrieval is conducted in order to find the old similar cases, and the old solutions of the retrieved cases are adapted to solve the new problem. In this paper, a systematic three-level adaptation is proposed to reduce the effort required of the users in choosing the suitable solution to solve their problem. An example is used to illustrate in detail the proposed approach.

Su-Field analysis, as one of the inventive problem solving tools, can be used to analyze and improve the efficacy of the technical system. Generally, the process of using Su-Field model to solve a specific inventive problem includes: building a Problem Model, mapping to a Generic Problem Model, finding a Generic Solution Model based on the corresponding inventive standard, and finally establishing and instantiating a Solution Model. As one of the most important phases of Su- Field analysis, the last step is normally implemented manually with the help of physical effects, which link generic technical functions with specific applications and systems. The physical effects compatible with the context of the specific problem should be chosen to assist the users to instantiate the Solution Model. However, the physical effects and the specific problems are built at different levels of abstraction, and it is difficult for the users to choose, that is, given a certain function, too many physical effects are chosen while with the detailed context of the problem, no physical effect is returned. This paper proposes a new way of representing knowledge: both inventive standards and physical effects are represented as the change of two states, that is, the couple of the problem standard and the solution standard for inventive standards, and the couple of two states before and after applying physical effects. Firstly, three ontologies, that is, Su-Field Model Ontology, Su-Field Analysis Ontology and Physical Effects Ontology, are built to describe the problems with different granularities, and then the IS (Inventive Standard) rule and PE (Physical Effect) rules are established respectively for two kinds of reasoning. Finally the ontology reasoning is launched to provide the heuristic physical effects for the users. A case is used to elaborate the whole process in detail.

The theory of inventive problem solving (TRIZ) was developed to solve inventive problems in different industrial fields. In recent decades, modern innovation theories and methods proposed several different knowledge sources, whose use requires extensive knowledge about different engineering domains. In order to facilitate the use of the TRIZ knowledge sources, this paper explores a heuristic TRIZ problem solving approach. Firstly, TRIZ users start solving inventive problem with the TRIZ knowledge source of their choice. Then other similar knowledge sources are used according to a calculation of semantic relatedness. Finally, heuristic solutions are returned by ontology reasoning on the knowledge sources. The case of a “Diving Fin” is used to show the heuristic TRIZ problem solving process in detail.