DataScope is InnScience's AI-powered search engine for finding research papers and patents.
With the massive increase in the number of scholarly or technological documents like scientific journal articles or patents that are published annually, finding the right tool that allows researchers and industry professionals alike to find the most relevant and up to date information as quickly and easily as possible couldn't be more crucial. One way that the literature review and research discovery process can be streamlined is through the use of boolean search queries and NLP-powered search tools, rather than simple keyword searches.
DataScope has the capability of generating advanced boolean/keyword search queries, scanning and filtering search results using a unique AI-powered similarity index and enables users to organize and export research paper or patent search results. But one important feature that enables search results to be narrowed down is DataScope's ability to generate advanced boolean queries based on a user description. By using DataScope's boolean query capability, one can save significant time by reducing the number of search results to a digestible quantity eliminating endless scrolling through thousands of results.
Unlike the use of general keyword-based searches, when boolean queries are applied in combination with natural language processing (NLP), relevant search results can be returned that include the most relevant documents. DataScope uses natural language processing NLP-based tools, which take into account language statistical and context properties allowing the meaning behind document texts and queries to be understood and interpreted [1]. Such an application can return results determined using semantic similarities further contributing to a more manageable set of relevant results that help with time-saving and workflow optimization.
A Boolean Query Comparison
To demonstrate the value of DataScope's boolean query generator, let's compare InnScience's DataScope with a GPT's ability to produce a boolean query that provides a manageable, relevant, and specific set of search results for a very niche area of research (Geochemistry of Hydrothermal Vent Fluids at Oceanic Spreading Centres).
Here is a detailed description on which the queries are based: Geochemical investigations of hydrothermal vent fluids from black smokers on the Southern East Pacific Rise to understand how fluid-rock reactions generate hydrothermal fluids with distinct geochemical signatures. Processing and measurement of high temperature and pressure submarine hydrothermal fluids using methods such as ion chromatography (IC), inductively coupled plasma mass spectrometry (ICP-MS), in-situ chemical speciation modelling (e.g., SUPCRT92, EQ3/6). Research in the date range of 2016 to 2025.
Figures 1 and 2 below show each of the queries generated by a GPT and DataScope, respectively, based on the above research description.
Figure 1: Boolean query generated by a GPT based on the above description.
Figure 2: Boolean query generated by InnScience's DataScope based on the same research description above.
Using each of the above boolean queries to perform a scholarly article search using DataScope and its access to the Google Scholar database, there are some clear differences in the search results returned. The main differences are in the number of results, and the quality or specificity of the results with respect to the input query.
Number of Results
Searching the same database using the two different queries, Figures 3 and 4 show that the GPT-generated query returned nearly 5 times as many results compared to the DataScope-generated query (168 vs. 37).
Figure 3: A screenshot from the InnScience platform using DataScope displaying the number of results (168) found using the boolean query generated by GPT.
Figure 4: A screenshot from the InnScience platform using DataScope displaying the number of results (37) returned using the boolean query generated by InnScience's DataScope.
Result Contents vs. Query Specifics
Figures 5 and 6 below display the search results found by the GPT-generated and DataScope-generated queries, respectively. When comparing the dates of documents returned in results, DataScope's query provided more constrained results that fall within the date range specified in the queries (2016-2025). Out of the results returned, 100% of those found by DataScope's query fell within the correct date range, whereas those found using the GPT-generated query only provided 38% that fell within the correct date range. This means that 62% of the GPT-query results were out of date, some as much as 23-32 years older than specified, whereas none of the DataScope results were out-of-date!
Figure 5: Search results found using the GPT-generated query showing many out-of-date documents.
Figure 6: Search results found in DataScope using the DataScope-generated query showing all documents within the specified date range. Also displayed in this screenshot are the unique Similarity Index scores generated by DataScope's NLP-powered semantic analyses.
The Results Speak for Themselves
Using InnScience's DataScope as a research tool it is possible to streamline the research and discovery process with its AI-driven tools. Advanced boolean query generation that can drastically reduce the number of results obtained and provide more relevant and specific documents, particularly compared to other GPT-based platforms, is one of the various benefits of using InnScience's DataScope. Whether searching for publications, projects, facilities, IP or other research assets and outputs on the largest scholarly/academic database, or official patent databases, DataScope has you covered. Skip the GPTs and try DataScope for a faster, more efficient, and systematic AI-powered research tool.
References
[1]
Borisova, N., Karashtranova, E., & Atanasova, I. (2025). The advances in natural language processing technology and its impact on modern society. International Journal of Electrical & Computer Engineering (2088-8708), 15(2).