IPC Prediction of Patent Documents Using Neural Network with Attention for Hierarchical Structure

International patent classifications (IPCs) are assigned to patent documents; however, since the procedure for assigning classifications is manually done by the patent examiner, it takes a lot of time and effort to select some IPCs from about 70,000 IPCs. Hence, some research has been conducted on patent classification with machine learning. However, patent documents are very voluminous, and learning with all the claims (the part describing the content of the patent) as input would run out of the necessary memory. Therefore, most of the existing methods learn by excluding some information, such as using only the first claim as input. In this study, we propose a model that considers the contents of all claims by extracting important information for input. We also propose a new decoder that considers the hierarchical structure of the IPC. Finally, we evaluate the model using an evaluation index that assumes the actual use of IPC selection for patent documents.

A Patent Analysis on Big Data Projects

Studies concerning Big Data patents have been published; however, research investigating Big Data projects is scarce. Therefore, the objective of this study was to conduct an exploratory analysis of a patent database to collect information about the characteristics of registered patents related to Big Data projects. We searched for patents related to Big Data projects in the Espacenet database on January 10, 2021, and identified 109 records.. The textual analysis detected three word classes interpreted as (i) a direction to cloud computing, (ii) optimization of solutions, and (iii) storage and data sharing structures. Our results also revealed emerging technologies such as Blockchain and the Internet of Things, which are utilized in Big Data project solutions. This observation demonstrates the importance that has been given to solutions that facilitate decision-making in an increasingly data-driven context. As a contribution, we understand that this study endorses a group of researchers that has been dedicated to academic research on patent documents.

Mycelium-Based Composites in Art, Architecture, and Interior Design: A Review

Mycelium-based composites (MBCs) have attracted growing attention due to their role in the development of eco-design methods. We concurrently analysed scientific publications, patent documents, and results of our own feasibility studies to identify the current design issues and technologies used. A literature inquiry in scientific and patent databases (WoS, Scopus, The Lens, Google Patents) pointed to 92 scientific publications and 212 patent documents. As a part of our own technological experiments, we have created several prototype products used in architectural interior design. Following the synthesis, these sources of knowledge can be concluded: 1. MBCs are inexpensive in production, ecological, and offer a high artistic value. Their weaknesses are insufficient load capacity, unfavourable water affinity, and unknown reliability. 2. The scientific literature shows that the material parameters of MBCs can be adjusted to certain needs, but there are almost infinite combinations: properties of the input biomaterials, characteristics of the fungi species, and possible parameters during the growth and subsequent processing of the MBCs. 3. The patent documents show the need for development: an effective method to increase the density and the search for technologies to obtain a more homogeneous internal structure of the composite material. 4. Our own experiments with the production of various everyday objects indicate that some disadvantages of MBCs can be considered advantages. Such an unexpected advantage is the interesting surface texture resulting from the natural inhomogeneity of the internal structure of MBCs, which can be controlled to some extent.

PatentNet: multi-label classification of patent documents using deep learning based language understanding

Patent classification is an expensive and time-consuming task that has conventionally been performed by domain experts. However, the increase in the number of filed patents and the complexity of the documents make the classification task challenging. The text used in patent documents is not always written in a way to efficiently convey knowledge. Moreover, patent classification is a multi-label classification task with a large number of labels, which makes the problem even more complicated. Hence, automating this expensive and laborious task is essential for assisting domain experts in managing patent documents, facilitating reliable search, retrieval, and further patent analysis tasks. Transfer learning and pre-trained language models have recently achieved state-of-the-art results in many Natural Language Processing tasks. In this work, we focus on investigating the effect of fine-tuning the pre-trained language models, namely, BERT, XLNet, RoBERTa, and ELECTRA, for the essential task of multi-label patent classification. We compare these models with the baseline deep-learning approaches used for patent classification. We use various word embeddings to enhance the performance of the baseline models. The publicly available USPTO-2M patent classification benchmark and M-patent datasets are used for conducting experiments. We conclude that fine-tuning the pre-trained language models on the patent text improves the multi-label patent classification performance. Our findings indicate that XLNet performs the best and achieves a new state-of-the-art classification performance with respect to precision, recall, F1 measure, as well as coverage error, and LRAP.

Reversal of Osseointegration as a Novel Perspective for the Removal of Failed Dental Implants: A Review of Five Patented Methods

Osseointegration is the basis of successful dental implantology and the foundation of cementless arthroplasty and the osseointegrated percutaneous prosthetic system. Osseointegration has been considered irreversible thus far. However, controlled heating or cooling of dental implants could selectively damage the bone at the bone–implant interface, causing the reversal of osseointegration or “osseodisintegration”. This review compares five methods for implant removal, published as patent documents between 2010 and 2018, which have not yet been discussed in the scientific literature. We describe these methods and evaluate their potential for reversing osseointegration. The five methods have several technical and methodological similarities: all methods include a handpiece, a connecting device for coronal access, and a controlling device, as well as the application of mechanical and/or thermal energy. The proposed method of quantifying the temperature with a sensor as the sole means for regulating the process seems inadequate. A database used in one of the methods, however, allows a more precise correlation between a selected implant and the energy needed for its removal, thus avoiding unnecessary trauma to the patient. A flapless, microinvasive, and bone-conserving approach for removing failed dental implants, facilitating successful reimplantation, would benefit dental implantology. These methods could be adapted to cementless medical implants and osseointegrated percutaneous prosthetics. However, for some of the methods discussed herein, further research may be necessary.

Mapping Scientific and Technological Production Related to the MYC Gene

In appropriate activation of c-MYC proto-oncogene contributes to the development of human cancers. Searches for therapies that target genes and proteins related to neoplastic phenotypes have become frequent. Therefore, inhibiting c-MYC expression has been the target for developing and testing multiple drugs and strategies for the treatment of various human cancers. This study aimed to map scientific and technological productions on the MYC gene at the Scielo, PubMed and Orbit Intelligence platforms between 2000 and 2019. The scientific prospecting revealed 1,259 articles. The most detected categories were: molecular biology, MYC mutations and those addressing the MYC as a drug target or therapeutic strategies. A progressive increase in the number of articles in this last category was found. Technological mapping detected 10,059 patent documents, with 20.2% granted. China and the USA were the largest filers, accounting for more than 40%. Biotechnology was the field with the highest number of patents. Biotechnology and the pharmaceutical sector predominated in the second half of the period investigated, both in scientific and technological prospecting. Our study points to a scientific and technological effort in the development of therapeutic strategies against cancer, in which MYC is among the main targets.

Technological Prospection with a Focus on Assistive Technology in Adapted Vehicles for People with Disabilities

This study aimed to review the production of patent documents related to assistive technology on adapted vehicles for people with disabilities. The technological prospection was carried out through a search on the Espacenet database, resulting in 277 patents whose data we organized into charts. The results show that the total number of patents filed in this area is small and with a limited increase over the years, maybe due to industrial secrecy. Twenty-eight countries registered patents. The Republic of China and the Korea Republic were the main depositors, while Brazil registered only one patent. They focused on people with a physical disability, so further projects to attend to other types of disability are also required. There is a necessity to increase government and private sectors investments for technological advances in this area.

What do patents tell us about microalgae in agriculture?

Microalgae have been used widely as a biological source for several industries, such as biofuel, pharmaceutical and food. Recently, the agricultural industry has also began using microalgae as an alternative source for sustainable products to replace agrochemicals. Due to the lack of scientific articles in this research area, the objective of this study was to search for applications of microalgae and to characterize its use in agriculture using the patent documents available in three patent databases, World Intellectual Property Organization (WIPO), European Patent Office (EPO) and Brazilian Institute of Industrial Property (INPI). The search was carried out using the keyword “microalgae” and applying the filter for International Patent Classification (IPC) code “A01N” which corresponds to patents related to agriculture and cultivation of microalgae. Our patent database search returned 669 documents and 132 patents were selected for the study based on their abstracts. The first patent was registered in 1982 and described the use of microalgae Chlorella extract as a plant growth promoter. After that, no patent was registered for 15 years. From 2005 to 2014, only seven patents were found. However, the scenario changed from 2015 when the number of patents increased mainly in the United States, China and Europe. The patent analysis showed several applications for microalgae in the agricultural sector, such as plant growth promotion, biofertilization, plant disease control, weed management, and post-harvest quality. This review confirmed the increasing interest in microalgae-derived products in agriculture and the value of using patent documents to assess innovative areas.

Disclosure and Subsequent Innovation: Evidence from the Patent Depository Library Program

How important is access to patent documents for subsequent innovation? We examine the expansion of the USPTO Patent Library system after 1975. Patent libraries provided access to patents before the Internet. We find that after patent library opening, local patenting increases by 8–20 percent relative to similar regions. Additional analyses suggest that disclosure of technical information drives this effect: inventors increasingly take up ideas from outside their region, and the effect is strongest in technologies where patents are more informative. We thus provide evidence that disclosure plays an important role in cumulative innovation. (JEL D83, K11, O31, O34, R11)