Systematic Literature Review using Dimensions Ai || Bibliometric Analysis and Visualization || Hindi

Dimension AI is positioned as a research database platform for running bibliometric analysis and systematic literature reviews (SLRs) by turning keyword searches into filtered, exportable datasets that can then be analyzed through built-in analytics and downstream visualization tools. The core workflow starts with searching for a research theme—such as “ECG signal analysis”—and then narrowing results using time windows, publication types, and other filters to reach a manageable set of records suitable for screening and meta-level synthesis.

After logging in and landing on the profile area, the process emphasizes search-driven identification of relevant literature. A keyword query can initially return very large result counts (the transcript cites figures in the hundreds of thousands to millions), but the platform supports tightening the scope by selecting a recent time range (e.g., last five years), applying topic/SDG-related constraints (the example filters toward “Good health and well-being”), and limiting by publication type (e.g., articles only). Additional narrowing can target specific classification groupings (the transcript references a “UGC CARE Group 2” filter) and exclude certain record categories (such as “open access” considerations), ultimately producing a final dataset size that is practical for analysis—around 11,000 documents in one filtered scenario and about 1,300 documents in the main export example.

With the filtered set ready, the platform’s export function becomes the bridge to bibliometric and SLR-style workflows. Export is capped at a maximum number of records per download (the transcript mentions a limit of 2,500), and exported files can be saved for later use in visualization and analysis environments. The transcript then demonstrates staying within Dimension AI’s analytics view: selecting an “analytical view” for the exported dataset and generating charts and summaries.

Key analytics shown include research category aggregation (presented as bar charts), publication overview metrics (total publications within the chosen time window, with counts and site-based indicators), and downloadable outputs such as PNG images and data files (CSV/PDF are also mentioned). The platform also supports author-level exploration through an “researcher” view, where contributors tied to the dataset can be listed and examined.

Beyond descriptive statistics, the transcript highlights network-style analysis. An author network view reveals how researchers are connected within the selected literature set, and the same dataset can be re-filtered by institution or author. For instance, typing an institution name (e.g., “Sikkim Manipal University”) is described as enabling direct navigation to the relevant analytical results, while author-specific filtering can isolate output associated with a particular researcher affiliation (the transcript cites an example tied to “AMU”).

Overall, Dimension AI is presented as a systematic, filter-first pipeline: search → screen-like filtering → exportable dataset → analytical dashboards and network views. The transcript also signals that deeper bibliometric techniques—such as visualization tools and additional analyses—can be handled in subsequent steps or companion videos, with PRISMA 2020 referenced as a framework for structuring identification, screening, selection, and final analysis stages in SLRs.